Search

Purpose: The relationship between external load and creatine-kinase (CK) response at the team/position or individual level using Global Positioning Systems (GPS) has been studied. This study aimed to compare GPS-derived and Football Movement Profile (FMP) –derived CK-prediction models for national-team soccer players. The second aim was to compare the performance of general and individualized CK prediction models. Methods: Four hundred forty-four national-team soccer players (under 15 [U15] to senior) were monitored during training sessions and matches using GPS. CK was measured every morning from whole blood. The players had 19.3 (18.1) individual GPS-CK pairs, resulting in a total of 8570 data points. Machine learning models were built using (1) GPS-derived or (2) FMP-based parameters or (3) the combination of the 2 to predict the following days' CK value. The performance of general and individual-specific prediction models was compared. The performance of the models was described by R2 and the root-mean-square error (RMSE, in units per liter for CK values). Results: The FMP model (R2 =.60, RMSE = 144.6 U/L) performed similarly to the GPS-based model (R2 =.62, RMSE = 141.2 U/L) and the combination of the 2 (R2 =.62, RMSE = 140.3 U/L). The prediction power of the general model was better on average (R2 =.57 vs R2 =.37) and for 73% of the players than the individualized model. Conclusions: The results suggest that FMP-based CK-prediction models perform similarly to those based on GPS-derived metrics. General machine learning models' prediction power was higher than those of the individual-specific models. These findings can be used to monitor postmatch recovery strategies and to optimize weekly training periodization. [ABSTRACT FROM AUTHOR]

Purpose: This study aimed to cross-validate a recently proposed equation for the prediction of maximal oxygen uptake ( V ˙ O 2 max) in cycling exercise by using the average power output normalized by the body mass from a 5-minute time trial (RPO5-min) as the independent variable. Further, the study aimed to update the predictive equation using Bayesian informative prior distributions and meta-analysis. Methods: On different days, 49 male cyclists performed an incremental graded exercise test until exhaustion and a 5-minute time trial on a stationary cycle ergometer. We compared the actual V ˙ O 2 max with the predicted value obtained from the RPO5-min, using a modified Bayesian Bland–Altman agreement analysis. In addition, this study updated the data on the linear regression between V ˙ O 2 max and RPO5-min, by incorporating information from a previous study as a Bayesian informative prior distribution or via meta-analysis. Results: On average, the predicted V ˙ O 2 max using RPO5-min underestimated the actual V ˙ O 2 max by −6.6 mL·kg–1·min–1 (95% credible interval, −8.6 to −4.7 mL·kg–1·min–1). The lower and upper 95% limits of agreement were −17.2 (−22.7 to −12.3) and 3.8 (−1.0 to 9.5) mL·kg–1·min–1, respectively. When the current study's data were analyzed using the previously published data as a Bayesian informative prior distribution, the accuracy of predicting sample means was found to be better when compared with the data combined via meta-analyses. Conclusions: The proposed equation presented systematic bias in our sample, in which the prediction underestimated the actual V ˙ O 2 max. We provide an updated equation using the previous one as the prior distribution, which could be generalized to a greater audience of cyclists. [ABSTRACT FROM AUTHOR]

We propose a new work design model, SMART work design, that identifies five higher order categories of work characteristics, including stimulating work characteristics (task variety, skill variety, information processing requirements, and problem‐solving requirements), mastery work characteristics (job feedback, feedback from others, and role clarity), autonomous work characteristics (decision‐making autonomy, timing autonomy, and method autonomy), relational work characteristics (social support, task significance, and beneficiary contact), and tolerable work characteristics (low levels of: role overload, work–home conflict, and role conflict). Higher order confirmatory factor analysis of working participants provided initial evidence of this structure (Study 1, N = 1107), which was replicated in an additional dataset (Study 2, time 1, N = 709). To provide further evidence, we examined Study 2 data across three waves (N = 573) to show that each higher order factor at time 1 predicted time 3 job satisfaction either directly or via the theorized time 2 mediators (challenge appraisals, work meaningfulness, fulfillment of relatedness needs, and activated negative affect). In Study 3 (N = 108), employees' scores on specific higher order variables correlated with leader ratings of performance in the expected ways. The SMART work design model provides a unique integrating and multidimensional theory of work design that extends beyond existing models. The model can be used to facilitate the synthesis of research knowledge and guide scholars and practitioners to diagnose and address contemporary work design challenges. [ABSTRACT FROM AUTHOR]

Background: Maximal lower-body strength can be assessed both dynamically and isometrically; however, the relationship between the changes in these 2 forms of strength following resistance training is not well understood. Purpose: To systematically review and analyze the effects of resistance training on changes in maximal dynamic (1-repetition-maximum back squat, deadlift, and power clean) and position-matched isometric strength (isometric midthigh pull and the isometric squat). In addition, individual-level data were used to quantify the agreement and relationship between changes in dynamic and isometric strength. Methods: Databases were systematically searched to identify eligible articles, and meta-analysis procedures were performed on the extracted data. The raw results from 4 studies were acquired, enabling bias and absolute reliability measures to be calculated using Bland–Altman test of agreement. Results: Eleven studies met the inclusion criteria, which resulted in 29 isometric–dynamic change comparisons. The overall pooled effect was 0.13 in favor of dynamic testing; however, the prediction interval ranged from g = −0.49 to 0.75. There was no evidence of bias (P =.825) between isometric and dynamic tests; however, the reliability coefficient was estimated to be 16%, and the coefficient of variation (%) was 109.27. Conclusions: As a range of future effects can be expected when comparing isometric to dynamic strength changes following resistance training, and limited proportionality exists between changes in these 2 strength qualities, there is strong evidence that isometric and dynamic strength represent separate neuromuscular domains. These findings can be used to inform strength-assessment models in athlete populations. [ABSTRACT FROM AUTHOR]

Purpose: To evaluate statistical models developed for predicting medal-winning performances for international swimming events and generate updated performance predictions for the Paris 2024 Olympic Games. Methods: The performance of 2 statistical models developed for predicting international swimming performances was evaluated. The first model employed linear regression and forecasting to examine performance trends among medal winners, finalists, and semifinalists over an 8-year period. A machine-learning algorithm was used to generate time predictions for each individual event for the Paris 2024 Olympic Games. The second model was a Bayesian framework and comprised an autoregressive term (the previous winning time), moving average (past 3 events), and covariates for stroke, gender, distance, and type of event (World Championships vs Olympic Games). To examine the accuracy of the predictions from both models, the mean absolute error was determined between the predicted times for the Budapest 2022 World Championships and the actual results from said championships. Results: The mean absolute error for prediction of swimming performances was 0.80% for the linear-regression machine-learning model and 0.85% for the Bayesian model. The predicted times and actual times from the Budapest 2022 World Championships were highly correlated (r =.99 for both approaches). Conclusions: These models, and associated predictions for swimming events at the Paris 2024 Olympic Games, provide an evidence-based performance framework for coaches, sport-science support staff, and athletes to develop and evaluate training plans, strategies, and tactics, as well as informing resource allocation to athletes, based on their potential for the Paris 2024 Olympic Games. [ABSTRACT FROM AUTHOR]

Background: Laboratory assessment of maximal oxygen uptake ( V ˙ O 2 max) is physically and mentally draining for the athlete and requires expensive laboratory equipment. Indirect measurement of V ˙ O 2 max could provide a practical alternative to laboratory testing. Purpose: To examine the relationship between the maximal power output (MPO) in an individualized 7 × 2-minute incremental test (INCR-test) and V ˙ O 2 max and to develop a regression equation to predict V ˙ O 2 max from MPO in female rowers. Methods: Twenty female club and Olympic rowers (development group) performed the INCR-test on a Concept2 rowing ergometer to determine V ˙ O 2 max and MPO. A linear regression analysis was used to develop a prediction of V ˙ O 2 max from MPO. Cross-validation analysis of the prediction equation was performed using an independent sample of 10 female rowers (validation group). Results: A high correlation coefficient (r =.94) was found between MPO and V ˙ O 2 max. The following prediction equation was developed: V ˙ O 2 max (mL·min−1) = 9.58 × MPO (W) + 958. No difference was found between the mean predicted V ˙ O 2 max in the INCR-test (3480 mL·min−1) and the measured V ˙ O 2 max (3530 mL·min−1). The standard error of estimate was 162 mL·min−1, and the percentage standard error of estimate was 4.6%. The prediction model only including MPO, determined during the INCR-test, explained 89% of the variability in V ˙ O 2 max. Conclusion: The INCR-test is a practical and accessible alternative to laboratory testing of V ˙ O 2 max. [ABSTRACT FROM AUTHOR]

Purpose: To investigate the indirect measurement of 1-repetition-maximum (1RM) free-weight half-squat in high-level sprinters using the load–velocity relationship. Methods: Half-squat load and velocity data from 11 elite sprinters were collected in 2 separate testing sessions. Approximately 24 hours prior to the first testing session, sprinters completed a fatiguing high-intensity training session consisting of running intervals, staircase exercises, and body-weight exercises. Prior to the second testing session, sprinters had rested at least 48 hours. Two different prediction models (multiple-point method, 2-point method) were used to estimate 1RM based on the load and either mean or peak concentric velocity data of submaximal lifts (40%–90% 1RM). The criterion validity of all methods was examined through intraclass correlation coefficients, coefficient of variation (CV%), Bland–Altman plots, and the SEM. Results: None of the estimations were significantly different from the actual 1RM. The multiple-point method showed higher intraclass correlation coefficients (.91 to.97), with CVs from 3.6% to 11.7% and SEMs from 5.4% to 10.6%. The 2-point method showed slightly lower intraclass correlation coefficients (.76 to.95), with CVs 1.4% to 17.5% and SEMs from 9.8% to 26.1%. Bland–Altman plots revealed a mean random bias in estimation of 1RM for both methods (mean and peak velocity) ranging from 1.06 to 13.79 kg. Conclusion: Velocity-based methods can be used to roughly estimate 1RM in elite sprinters in the rested and fatigued conditions. However, all methods showed variations that limit their applicability for accurate load prescription for individual athletes. [ABSTRACT FROM AUTHOR]

Are employees more likely to identify with their organization and engage in helping behaviors on its behalf if the coworkers who make up their social networks identify and help the organization? We draw on social information processing theory, social learning theory, and research on diffusion of innovation to develop a model that predicts how relational (advice ties) and structural (structural equivalence) factors predict organizational identification and organizational citizenship. We argue that OI is neither a matter of individual perceptions generated in isolation, nor a simple function of the number of ties that an employee has, but that it is instead an outcome of relations and interactions with coworkers, as reflected in the social structure of the employing organization. We tested our model by conducting a social network study of employees in a construction company. Our findings show that (1) employees tend to have similar levels of OI to the OI of their advice ties; (2) employees who are structurally equivalent tend to demonstrate similar levels of OCB; (3) similarity in OI mediates the relationship between advice ties and similarity in OCB; and (4) advice ties with high and moderate levels of OI are most strongly associated with similarity in OCB. In a supplementary analysis we show that the social influence model that we tested offers a better explanation for the role that the social context plays in shaping OI than does a "more ties, more OI" model. We discuss implications for theory and practice. [ABSTRACT FROM AUTHOR]

Objective. The number of cancer survivors is increasing, and the high prevalence of frailty not only reduces quality of life but also affects the treatment of cancer patients. This study aimed to identify the prevalence and risk factors of frailty in cancer patients and to construct a nomogram to predict the probability of frailty. Methods. Nine hundred fifty‐eight cancer patients were included in this retrospective study, randomly divided into a development set (n = 680) and a validation set (n = 278). Frailty was assessed using the Tilburg frailty indicator (TFI). Social support, medical coping styles, and psychological distress were assessed by the Social Support Self‐Rating Scale (SSRS), Medical Coping Modes Questionnaire (MCMQ), and distress thermometer (DT), respectively. Results. The prevalence of frailty in cancer patients was 45.93%. Cancer patients who exercised regularly, ate a balanced diet, and actively coped with diseases were less likely to become frail. The risk factors for frailty identified by a multivariate analysis were parenteral nutrition, advanced TNM staging, vegetarian diet, unemployment, psychological distress ≥4, low physical activity, and negative coping styles. These risk factors were used to construct a nomogram, and the C‐index, calibration curve, and decision curve analysis (DCA) were used to assess the performance of the nomogram. The C‐index was 0.762, and the calibration curve showed satisfactory coherence. The net benefit of the nomogram was better between threshold probabilities of 17%–96% in DCA. Conclusion. Special focus needs to be placed on frail cancer patients due to their high prevalence and severe outcomes. Clinical medical workers could use this nomogram to identify high‐risk patients and intervene early to prevent frailty. [ABSTRACT FROM AUTHOR]

Objective: Cognitive decline is often considered an inevitable aspect of aging; however, recent research has identified a subset of older adults known as "superagers" who maintain cognitive abilities comparable to those of younger individuals. Investigating the neurobiological characteristics associated with superior cognitive function in superagers is essential for understanding "successful aging." Evidence suggests that the gut microbiome plays a key role in brain function, forming a bidirectional communication network known as the microbiome-gut-brain axis. Alterations in the gut microbiome have been linked to cognitive aging markers such as oxidative stress and inflammation. This study aims to investigate the unique patterns of the gut microbiome in superagers and to develop machine learning-based predictive models to differentiate superagers from typical agers. Methods: We recruited 161 cognitively unimpaired, community-dwelling volunteers aged 60 years or from dementia prevention centers in Seoul, South Korea. After applying inclusion and exclusion criteria, 115 participants were included in the study. Following the removal of microbiome data outliers, 102 participants, comprising 57 superagers and 45 typical agers, were finally analyzed. Superagers were defined based on memory performance at or above average normative values of middle-aged adults. Gut microbiome data were collected from stool samples, and microbial DNA was extracted and sequenced. Relative abundances of bacterial genera were used as features for model development. We employed the LightGBM algorithm to build predictive models and utilized SHAP analysis for feature importance and interpretability. Results: The predictive model achieved an AUC of 0.832 and accuracy of 0.764 in the training dataset, and an AUC of 0.861 and accuracy of 0.762 in the test dataset. Significant microbiome features for distinguishing superagers included Alistipes, PAC001137_g, PAC001138_g, Leuconostoc, and PAC001115_g. SHAP analysis revealed that higher abundances of certain genera, such as PAC001138_g and PAC001115_g, positively influenced the likelihood of being classified as superagers. Conclusion: Our findings demonstrate the machine learning-based predictive models using gut-microbiome features can differentiate superagers from typical agers with a reasonable performance. [ABSTRACT FROM AUTHOR]

Background: 3-Hydroxybutyrate, also called β-hydroxybutyrate, is a significant constituent of ketone bodies. Previous observational and experimental studies have suggested that ketogenic diet, especially 3-hydroxybutyrate, may have a protective effect against cardiovascular disease. However, the relationship between ketone bodies, especially 3-hydroxybutyrate, and aortic dissection remains uncertain. Materials and methods: Publicly accessible data from genome-wide association study (GWAS) was utilized to obtain information on ketone bodies, including 3-hydroxybutyrate, acetoacetate and acetone as exposure respectively, while GWAS data on aortic dissection was used as outcome. Subsequently, two-sample Mendelian randomization (MR) analysis was conducted to examine the potential relationship between ketone bodies and aortic dissection. Then, reverse and multivariate Mendelian randomization analyses were performed. Additionally, sensitivity tests were conducted to assess the robustness of MR study. Results: The inverse-variance weighted (IVW) method of Mendelian randomization analysis of gene prediction observed a negative correlation between 3-hydroxybutyrate and risk of aortic dissection (OR 0.147, 95% CI 0.053–0.410). Furthermore, consistent findings were obtained through the implementation of the weighted median, simple mode, Mendelian randomization-Egger (MR-Egger), and weighted mode methods. After adjusting acetoacetate (OR 0.143, 95% CI 0.023-0.900) or acetone (OR 0.100, 95% CI 0.025–0.398), MR analysis of gene prediction still observed a negative correlation between 3-hydroxybutyrate and risk of aortic dissection. No indications of heterogeneity or pleiotropy among the SNPs were detected. Conclusion: The findings from the MR analysis demonstrated that genetically predicted 3-hydroxybutyrate exhibits a protective effect against aortic dissection. [ABSTRACT FROM AUTHOR]

Background: Adverse events of the fractured vertebra (AEFV) post-percutaneous kyphoplasty (PKP) can lead to recurrent pain and neurological damage, which considerably affect the prognosis of patients and the quality of life. This study aimed to analyze the risk factors of AEFV and develop and select the optimal risk prediction model for AEFV to provide guidance for the prevention of this condition and enhancement of clinical outcomes. Methods: This work included 383 patients with primary osteoporotic vertebral compression fracture (OVCF) who underwent PKP. The patients were grouped based on the occurrence of AEFV postsurgery, and data were collected. Group comparisons and correlation analysis were conducted to identify potential risk factors, which were then included in the five prediction models. The performance indicators served as basis for the selection of the best model. Results: Multivariate logistic regression analysis revealed the following independent risk factors for AEFV: kissing spine (odds ratio (OR) = 8.47, 95% confidence interval (CI) 1.46–49.02), high paravertebral muscle fat infiltration grade (OR = 29.19, 95% CI 4.83–176.04), vertebral body computed tomography value (OR = 0.02, 95% CI 0.003–0.13, P < 0.001), and large Cobb change (OR = 5.31, 95% CI 1.77–15.77). The support vector machine (SVM) model exhibited the best performance in the prediction of the risk of AEFV. Conclusion: Four independent risk factors were identified of AEFV, and five risk prediction models that can aid clinicians in the accurate identification of high-risk patients and prediction of the occurrence of AEFV were developed. [ABSTRACT FROM AUTHOR]

Background: This study aims to evaluate the integration of optical coherence tomography (OCT) and peripheral blood immune indicators for predicting oral cancer prognosis by artificial intelligence. Methods: In this study, we examined patients undergoing radical oral cancer resection and explored inherent relationships among clinical data, OCT images, and peripheral immune indicators for oral cancer prognosis. We firstly built a peripheral blood immune indicator-guided deep learning feature representation method for OCT images, and further integrated a multi-view prognostic radiomics model incorporating feature selection and logistic modeling. Thus, we can assess the prognostic impact of each indicator on oral cancer by quantifying OCT features. Results: We collected 289 oral mucosal samples from 68 patients, yielding 1,445 OCT images. Using our deep radiomics-based prognosis model, it achieved excellent discrimination for oral cancer prognosis with the area under the receiver operating characteristic curve (AUC) of 0.886, identifying systemic immune-inflammation index (SII) as the most informative feature for prognosis prediction. Additionally, the deep learning model also performed excellent results with 85.26% accuracy and 0.86 AUC in classifying the SII risk. Conclusions: Our study effectively merged OCT imaging with peripheral blood immune indicators to create a deep learning-based model for inflammatory risk prediction in oral cancer. Additionally, we constructed a comprehensive multi-view radiomics model that utilizes deep learning features for accurate prognosis prediction. The study highlighted the significance of the SII as a crucial indicator for evaluating patient outcomes, corroborating our clinical statistical analyses. This integration underscores the potential of combining imaging and blood indicators in clinical decision-making. Trial registration: The clinical trial associated with this study was prospectively registered in the Chinese Clinical Trial Registry with the trial registration number (TRN) ChiCTR2200064861. The registration was completed on 2021. [ABSTRACT FROM AUTHOR]

Background: Mind--body treatments can improve coping mechanisms to deal with pain, improve the quality of life of patients with fibromyalgia syndrome (FMS), and reduce perceived pain in some cases. However, responses to these treatments are highly variable, the mechanisms underpinning them remain unclear, and reliable predictors of treatment response are lacking. We employed resting-state blood oxygen level-dependent (rsBOLD) functional magnetic resonance imaging (fMRI) to examine changes in brain functional connectivity (FC) following mind--body treatment that may relate to and predict pain relief. Methods: We recruited patients with FMS who underwent either mindfulnessbased stress reduction (MBSR; n = 18) or a psychoeducational program (FibroQoL; n = 22) and a treatment-as-usual FMS group (TAU; n = 18). We collected rsBOLD data, alongside subjective pain, anxiety, depression, and catastrophizing measures prior to and following treatments. We examined behavioral changes and FC changes in the salience network (SN) and sensorimotor network (SMN) and performed regression analyses to identify predictors for treatment response. Results: The MBSR and FibroQoL groups experienced significant reductions in pain catastrophizing. After treatment, the FC of the sensorimotor cortex with the rest of the SMN became significantly reduced in the MBSR group compared to the TAU group. The FC between the SN and the SMN at baseline was negatively correlated with pain reductions following MBSR but positively correlated with pain reductions in the FibroQoL group. These results yielded large to very large effect sizes. Following MBSR, only for those patients with lower baseline SMN-SN FC, minutes of mindfulness practice were positively associated with clinical improvement (small to medium effect size). Conclusions: Different mind--body treatments are underpinned by discrete brain networks. Measures of the functional interplay between SN and SMN have the potential as predictors of mind--body treatment response in patients with FMS. [ABSTRACT FROM AUTHOR]

Health disparities are ongoing challenges in the United States, with one of the underlying factors being the variation in health risk behaviors across different socioeconomic and demographic communities. In this cross-sectional ecological study, utilizing data from 26,781 US zip codes provided by the Centers for Disease Control and Prevention (CDC), we conducted generalized linear model regressions to explore the associations between race, poverty, education, and urban/rural status in areas. Our analysis indicated that areas with a higher prevalence of Black population and higher poverty levels are associated with a more significant number of unique health risk behaviors. Conversely, other racial groups were associated with fewer unique health risk behaviors, although they may still engage in higher levels of specific behaviors. These insights underscore the need for tailored public health strategies to address the disparities in health risk behaviors across different sub-populations. Additionally, we employed Finite Mixture Modeling (FMM) to identify distinct sub-populations in the United States based on health risk behaviors, specifically binge drinking, smoking, sleep deprivation, and physical inactivity, which are essential risk factors for chronic diseases such as cancer, diabetes, and cardiovascular diseases. We identified four clusters and conducted an in-depth analysis of the demographics and socioeconomic characteristics of the identified clusters. Our findings reveal that one cluster, characterized by the worst health risk behaviors, exhibited the highest poverty rate, the least favorable health insurance profile, and the lowest frequencies of post-secondary education attainment compared to the other three clusters. This cluster, therefore, warrants targeted public health interventions. [ABSTRACT FROM AUTHOR]

This study investigated how equipment and sex affect the prediction accuracy of the maximum number of repetitions performed to failure (RTF) using the fastest mean velocity of the set (MVfastest). Sixteen men and twelve women completed four sessions (two using free-weight equipment and two sessions using the Smith machine). Each session involved three sets of repetitions to failure against the 65%, 75%, and 85% of the one-repetition maximum, interspersed by 10-min of rest. The goodness-of-fit of the individualized RTF-MVfastest relationships was comparable between both equipment types and sexes (P ≥0.510). Moreover, there were not significant differences in the MVfastest associated with RTF between equipment types (P ≥0.258). However, the MVfastest associated with RTF was higher for men than for women in repetitions 6 to 15 (P ≤0.043; ES≥0.69). In addition, the absolute errors when predicting RTF showed no significant differences between equipment types and loads (P <0.444). Specifically, these RTF estimates were within an acceptable range for men (<2 repetitions), but not for women (≥2 repetitions) (main effect of sex: P ≤0.018; ES≥0.58). These findings suggest that individualized RTF-MVfastest equations estimate the RTF with an acceptable precision in men during bench press exercises in both equipment types but exhibit lower precision for women. [ABSTRACT FROM AUTHOR]

Background: In South Africa, between 1966 and 2014, there were three kidney transplant eras defined by evolving access to certain immunosuppressive therapies defined as Pre-CYA (before availability of cyclosporine), CYA (when cyclosporine became available), and New-Gen (availability of tacrolimus and mycophenolic acid). As such, factors influencing kidney graft failure may vary across these eras. Therefore, evaluating the consistency and reproducibility of models developed to study these variations using machine learning (ML) algorithms could enhance our understanding of post-transplant graft survival dynamics across these three eras. Methods: This study explored the effectiveness of nine ML algorithms in predicting 10-year graft survival across the three eras. We developed and internally validated these algorithms using data spanning the specified eras. The predictive performance of these algorithms was assessed using the area under the curve (AUC) of the receiver operating characteristics curve (ROC), supported by other evaluation metrics. We employed local interpretable model-agnostic explanations to provide detailed interpretations of individual model predictions and used permutation importance to assess global feature importance across each era. Results: Overall, the proportion of graft failure decreased from 41.5% in the Pre-CYA era to 15.1% in the New-Gen era. Our best-performing model across the three eras demonstrated high predictive accuracy. Notably, the ensemble models, particularly the Extra Trees model, emerged as standout performers, consistently achieving high AUC scores of 0.95, 0.95, and 0.97 across the eras. This indicates that the models achieved high consistency and reproducibility in predicting graft survival outcomes. Among the features evaluated, recipient age and donor age were the only features consistently influencing graft failure throughout these eras, while features such as glomerular filtration rate and recipient ethnicity showed high importance in specific eras, resulting in relatively poor historical transportability of the best model. Conclusions: Our study emphasises the significance of analysing post-kidney transplant outcomes and identifying era-specific factors mitigating graft failure. The proposed framework can serve as a foundation for future research and assist physicians in identifying patients at risk of graft failure. [ABSTRACT FROM AUTHOR]

Pharmacotherapy guidelines for type 2 diabetes (T2D) emphasize patient-centered care, but applying this approach effectively in outpatient practice remains challenging. Data-driven treatment optimization approaches could enhance individualized T2D management, but current approaches cannot account for drug-specific and dose-dependent variations in safety and efficacy. We developed and evaluated an AI Drug mix and dose Advisor (AIDA) for glycemic management, using electronic medical records from 107,854 T2D patients in the SingHealth Diabetes Registry. Given a patient's medical profile, AIDA leverages a predict-then-optimize approach to identify the minimal drug mix and dose changes required to optimize glycemic control, subject to clinical knowledge-based guidelines. On unseen data from large internal, external, and temporal validation sets, AIDA recommendations were estimated to improve post-visit glycated hemoglobin (HbA1c) by an average of 0.40–0.68% over standard of care (P < 0.0001). In qualitative evaluations on 60 diverse cases by a panel of three endocrinologists, AIDA recommendations were mostly rated as reasonable and precise. Finally, AIDA's ability to account for drug-dose specifics offered several advantages over competing methods, including greater consistency with practice preferences and clinical guidelines for practical but effective options, indication-based treatments, and renal dosing. As AIDA provides drug-dose recommendations to improve outcomes for individual T2D patients, it could be used for clinical decision support at point-of-care, especially in resource-limited settings. [ABSTRACT FROM AUTHOR]

Background: Haemoglobin variation (ΔHb) induced by fluid transfer through the intestitium has been proposed as a useful tool for detecting hydrostatic pulmonary oedema (HPO). However, its use in the emergency department (ED) setting still needs to be determined. Methods: In this observational retrospective monocentric study, ED patients admitted for acute dyspnoea were enrolled. Hb values were recorded both at ED presentation (T0) and after 4 to 8 h (T1). ΔHb between T1 and T0 (ΔHbT1-T0) was calculated as absolute and relative value. Two investigators, unaware of Hb values, defined the cause of dyspnoea as HPO and non-HPO. ΔHbT1-T0 ability to detect HPO was evaluated. A machine learning approach was used to develop a predictive tool for HPO, by considering the ability of ΔHb as covariate, together with baseline patient characteristics. Results: Seven-hundred-and-six dyspnoeic patients (203 HPO and 503 non-HPO) were enrolled over 19 months. Hb levels were significantly different between HPO and non-HPO patients both at T0 and T1 (p < 0.001). ΔHbT1-T0 were more pronounced in HPO than non-HPO patients, both as relative (-8.2 [-11.2 to -5.6] vs. 0.6 [-2.1 to 3.3] %) and absolute (-1.0 [-1.4 to -0.8] vs. 0.1 [-0.3 to 0.4] g/dL) values (p < 0.001). A relative ΔHbT1-T0 of -5% detected HPO with an area under the receiver operating characteristic curve (AUROC) of 0.901 [0.896–0.906]. Among the considered models, Gradient Boosting Machine showed excellent predictive ability in identifying HPO patients and was used to create a web-based application. ΔHbT1-T0 was confirmed as the most important covariate for HPO prediction. Conclusions: ΔHbT1-T0 in patients admitted for acute dyspnoea reliably identifies HPO in the ED setting. The machine learning predictive tool may represent a performing and clinically handy tool for confirming HPO. [ABSTRACT FROM AUTHOR]

Background: Prostate cancer is well known as the commonest cancer in men and the second leading cause of cancer-related death. CT, MRI and bone scintigraphy are considered the commonly widely used imaging diagnostic tools for detection, staging and follow-up of prostate cancer. Prostate-specific membrane antigen (PSMA) is a membrane glycoprotein, that can be concentrated in prostate cancer cells up to 100 times higher than in normal cells. PSMA-targeted imaging modalities have now proven their efficacy in diagnosis, staging and follow-up of prostate cancer. The use of 68Ga PSMA PET-CT has efficiently improved the detection of loco-regional and metastatic disease. 68Ga PSMA PET-CT also has an effective role in the primary diagnosis, staging, and detecting biochemical recurrence after curative treatment and in metastasis-targeted therapy. This work aims to review the role of 68Ga PSMA PET-CT in anatomical staging of prostate cancer in correlation with histopathological staging. Results: Zonal correlation between 68Ga PSMA findings and biopsy results showed sensitivity ranging between 76.9 and 90.6% and specificity ranging from 85.7 to 100%. There was high significant correlation between the SUVmax uptake and the biopsy results, between the SUVmax uptake and the local staging as well as between the Gleason score and 68Ga PSMA PET/CT findings. Conclusions: 68Ga PSMA PET/CT is a highly promising imaging modality with an effective role in detection of prostate cancer showing high sensitivity and specificity in prediction of zonal histopathological results and loco-regional Gleason score staging with significant positive correlation between the SUV uptake results, Gleason score and the PSA levels. [ABSTRACT FROM AUTHOR]

Background: Labor pain is the nastiest conceivable pain women are involved in during labor and delivery. In this way, the end of labor pain is frequently outlined by myths and equivocalness. Hence, giving a compelling absence of pain in labor remained a challenge specifically in developing countries including Ethiopia. Objectives: This systematic review and meta-analysis aimed to explore the pooled prevalence of labor analgesia and associated factors to pregnant women in Ethiopia. Design: A systematic review and meta-analysis were utilized in agreement with the Preferred Reporting Items for Systematic Reviews. Data Sources: PubMed/Medline, SCOPUS, EMBASE, Web of Science, Google Scholars, and the Cochrane Library and supplemented it with manual were deliberately looked at until January 1–30, 2024. Methods: Two authors independently extricated all principal information utilizing standardized data extraction designs, and the analysis was done utilizing STATA version 17. Heterogeneity over the studies was evaluated utilizing I 2 measurement. The funnel plot and Egger's weighted regression tests were utilized to assess subjective and objective publication biases respectively. Also, the pooled effect of labor pain management and the associations were evaluated utilizing a random-effects model. Results: The general pooled prevalence of labor analgesia in the present study was 23.3% with a (95% confidence interval (CI): 13.5, 33.1). Maternal age (odds ratio (OR): 1.91; 95% CI: 1.11, 2.77), parity of the mother (OR: 0.28; 95% CI: 0.06, 0.63), history of pregnancy misfortune (OR: 0.12; 95% CI: 0.11, 0.36), length of labor (OR: 2.09; 95% CI: 1.06, 3.13), and awareness about labor analgesia (OR: 1.91; 95% CI: 0.34, 3.49) were significantly related with labor analgesia among pregnant women in Ethiopia. Conclusions: The generally pooled prevalence of labor analgesia among pregnant women in Ethiopia was low. Maternal age, parity of the mother, history of pregnancy loss, length of labor, and awareness of labor analgesia were factors influencing labor analgesia among pregnant women in Ethiopia. This finding proposes exceptional consideration to make laboring mothers free of pain by scaling up the strategies and utilizing labor pain administration in a way that universally recognized standards are met. Registration number: PROSPERO CRD: 42024525636 [ABSTRACT FROM AUTHOR]

Transferring and replicating predictive algorithms across healthcare systems constitutes a unique yet crucial challenge that needs to be addressed to enable the widespread adoption of machine learning in healthcare. In this study, we explored the impact of important differences across healthcare systems and the associated Electronic Health Records (EHRs) on machine-learning algorithms to predict mental health crises, up to 28 days in advance. We evaluated both the transferability and replicability of such machine learning models, and for this purpose, we trained six models using features and methods developed on EHR data from the Birmingham and Solihull Mental Health NHS Foundation Trust in the UK. These machine learning models were then used to predict the mental health crises of 2907 patients seen at the Rush University System for Health in the US between 2018 and 2020. The best one was trained on a combination of US-specific structured features and frequency features from anonymized patient notes and achieved an AUROC of 0.837. A model with comparable performance, originally trained using UK structured data, was transferred and then tuned using US data, achieving an AUROC of 0.826. Our findings establish the feasibility of transferring and replicating machine learning models to predict mental health crises across diverse hospital systems. [ABSTRACT FROM AUTHOR]

The application of artificial intelligence (AI) algorithms to 12-lead electrocardiogram (ECG) provides promising age prediction models. We explored whether the gap between the pre-procedural AI-ECG age and chronological age can predict atrial fibrillation (AF) recurrence after catheter ablation. We validated a pre-trained residual network-based model for age prediction on four multinational datasets. Then we estimated AI-ECG age using a pre-procedural sinus rhythm ECG among individuals on anti-arrhythmic drugs who underwent de-novo AF catheter ablation from two independent AF ablation cohorts. We categorized the AI-ECG age gap based on the mean absolute error of the AI-ECG age gap obtained from four model validation datasets; aged-ECG (≥10 years) and normal ECG age (<10 years) groups. In the two AF ablation cohorts, aged-ECG was associated with a significantly increased risk of AF recurrence compared to the normal ECG age group. These associations were independent of chronological age or left atrial diameter. In summary, a pre-procedural AI-ECG age has a prognostic value for AF recurrence after catheter ablation. [ABSTRACT FROM AUTHOR]

Background: Lacunes, a characteristic feature of cerebral small vessel disease (CSVD), are critical public health concerns, especially in the aging population. Traditional neuroimaging techniques often fall short in early lacune detection, prompting the need for more precise predictive models. Methods: In this retrospective study, 587 patients from the Neurology Department of the Affiliated Hospital of Hebei University who underwent cranial MRI were assessed. A nomogram for predicting lacune incidence was developed using LASSO regression and binary logistic regression analysis for variable selection. The nomogram's performance was quantitatively assessed using AUC-ROC, calibration plots, and decision curve analysis (DCA) in both training (n = 412) and testing (n = 175) cohorts. Results: Independent predictors identified included age, gender, history of stroke, carotid atherosclerosis, hypertension, creatinine, and homocysteine levels. The nomogram showed an AUC-ROC of 0.814 (95% CI: 0.791-0.870) for the training set and 0.805 (95% CI: 0.782-0.843) for the testing set. Calibration and DCA corroborated the model's clinical value. Conclusion: This study introduces a clinically useful nomogram, derived from binary logistic regression, that significantly enhances the prediction of lacunes in patients undergoing brain MRI for various indications, potentially advancing early diagnosis and intervention. While promising, its retrospective design and single-center context are limitations that warrant further research, including multi-center validation. [ABSTRACT FROM AUTHOR]

Objectives: Temporomandibular disorders (TMDs) have a relatively high prevalence among university students. This study aimed to identify independent risk factors for TMD in university students and develop an effective risk prediction model. Methods: This study included 1,122 university students from four universities in Changchun City, Jilin Province, as subjects. Predictive factors were screened by using the least absolute shrinkage and selection operator (LASSO) regression and the machine learning Boruta algorithm in the training cohort. A multifactorial logistic regression analysis was used to construct a TMD risk prediction model. Internal validation of the model was conducted via bootstrap resampling, and an external validation cohort comprised 205 university students undergoing oral examinations at the Stomatological Hospital of Jilin University. Results: The prevalence of TMD among university students was 44.30%. Ten predictive factors were included in the model, comprising gender, facial cold stimulation, unilateral chewing, biting hard or resilient foods, clenching teeth, grinding teeth, excessive mouth opening, malocclusion, stress, and anxiety. The model demonstrated good predictive ability with area under the receiver operating characteristic curve (AUC) values of 0.853, 0.838, and 0.821 in the training cohort, internal validation cohort, and external validation cohort, respectively. The calibration curves demonstrated that the predicted results were consistent with the actual results, and the decision curve analysis (DCA) indicated the model's high clinical utility. Conclusions: An online nomogram of TMD in university students with good predictive performance was constructed, which can effectively predict the risk of TMD in university students. The model provides a useful tool for the early identification and treatment of TMDs in university students, helping clinicians to predict the probability of TMDs in each patient, thus providing more personalized and accurate treatment decisions for patients. [ABSTRACT FROM AUTHOR]

Objectives To assess whether there are identifiable subgroups of disease activity trajectory in a population of JDM patients—followed throughout childhood and into adulthood—and determine factors that predict those trajectory groupings. Methods This is a retrospective, longitudinal inception cohort of patients with idiopathic inflammatory myopathies, largely JDM. We sought to identify baseline factors that predict membership into different groups (latent classes) of disease activity trajectory. Results A total of 172 patients (64% females), with median age at diagnosis of 7.7 years, were analysed. We studied 4725 visits (1471 patient-years). We identified three latent classes of longitudinal disease activity, as measured by the modified DAS (DASm), with distinct class trajectories predicted by DASm at baseline, and by the changes of DASm from either baseline to 3 months or baseline to 6 months (early response to therapy). In the analysis in which DASm at baseline and the changes of DASm from baseline to 6 months are included as predictors, Class 1 (10%) has persistently high disease activity, Class 2 (34%) is characterized by moderate disease activity and Class 3 (56%) is characterized by individuals with a high early disease activity but an apparently good response to treatment and long-term low disease activity. Conclusion High early disease activity, and treatment resistance in the first few months, predict a more chronic longitudinal course of JDM. [ABSTRACT FROM AUTHOR]

Background: While home is frequently expressed as the favorite place of death (PoD) among terminally ill cancer patients, various factors affect the fulfillment of this wish. The determinants of the PoD of cancer patients in countries without healthcare system-integrated palliative and supportive care have not been studied before. This study aimed at identifying the predictors of the PoD of patients who suffer from advanced cancer by developing a reliable predictive model among who received home-based palliative care in Iran as a representative of the countries with isolated provision of palliative care services. Methods: In a cross-sectional study, electronic records of 4083 advanced cancer patients enrolled in the Iranian Cancer Control Center (MACSA) palliative homecare program, who died between February 2018 and February 2020 were retrieved. Multivariable binary logistic regression analysis as well as subgroup analyses (location, sex, marital status, and tumor topography) was performed to identify the predictors of PoD. Results: Of the 2398 cases included (mean age (SD) = 64.17 (14.45) year, 1269 (%52.9) male), 1216 (50.7%) patients died at home. Older age, presence and intensity of medical homecare in the last two weeks and registration in the Tehran site of the program were associated with dying at home (P < 0.05). Gynecological or hematological cancers, presence and intensity of the calls received from the remote palliative care unit in the last two weeks were predictors of death at the hospital (p < 0.05). The model was internally and externally validated (AUC = 0.723 (95% CI = 0.702–0.745; P < 0.001) and AUC = 0.697 (95% CI = 0.631–0.763; P < 0.001) respectively). Conclusion: Our model highlights the demographic, illness-related and environmental determinants of the PoD in communities with patchy provision of palliative care. It also urges policymakers and service providers to identify and take the local determinant of the place of death into account to match the goals of palliative and supportive services with the patient preferences. [ABSTRACT FROM AUTHOR]

Background: Severe trauma is the leading cause of disability and mortality in the patients under 35 years of age. Surgical site infections (SSI) represent a significant complication in this patient population. However, they are often inadequately investigated, potentially impacting the quality of patient outcomes. The aim of this study was to investigate the epidemiology of SSI and risk factors in severe trauma patients. Methods: We conducted a multicenter retrospective cohort study screening the severe trauma patients (STP) admitted to two intensive care units of an academic institution in Marseille between years2018 and 2019. Those who underwent orthopedic or spinal surgery within 5 days after admission were included and classified into two groups according to the occurrence of SSI (defined by the Centers for Disease Control (CDC) international diagnostic criteria) or not. Our secondary goal was to evaluate STP survival at 48 months, risk factors for SSI and microbiological features of SSI. Results: Forty-seven (23%) out of 207 STP developed an SSI. Mortality at 48-months did not differ between SSI and non-SSI patients (12.7% vs. 10.0%; p = 0.59). The fractures of 22 (47%) severe trauma patients with SSI were classified as Cauchoix 3 grade and 18 (38%) SSI were associated with the need for external fixators. Thirty (64%) severe trauma patients with SSI had polymicrobial infection, including 34 (72%) due to Gram-positive cocci. Empirical antibiotic therapy was effective in 31 (66%) cases. Multivariate analysis revealed that risk factors such as low hemoglobin, arterial oxygenation levels, hyperlactatemia, high serum creatinine and glycemia, and Cauchoix 3 grade on the day of surgery were associated with SSI in severe trauma patients. The generated predictive model showed a good prognosis performance with an AUC of 0.80 [0.73–0.88] and a high NPV of 95.9 [88.6–98.5] %. Conclusions: Our study found a high rate of SSI in severe trauma patients, although SSI was not associated with 48-month mortality. Several modifiable risk factors for SSI may be effectively managed through enhanced perioperative monitoring and the implementation of a patient blood management strategy. [ABSTRACT FROM AUTHOR]

Background: Machine learning (ML) is extensively employed for forecasting the outcome of various illnesses. The objective of the study was to develop ML based classifiers using a stacking ensemble strategy to predict the Japanese Orthopedic Association (JOA) recovery rate for patients with degenerative cervical myelopathy (DCM). Methods: A total of 672 patients with DCM were included in the study and labeled with JOA recovery rate by 1-year follow-up. All data were collected during 2012–2023 and were randomly divided into training and testing (8:2) sub-datasets. A total of 91 initial ML classifiers were developed, and the top 3 initial classifiers with the best performance were further stacked into an ensemble classifier with a supported vector machine (SVM) classifier. The area under the curve (AUC) was the main indicator to assess the prediction performance of all classifiers. The primary predicted outcome was the JOA recovery rate. Results: By applying an ensemble learning strategy (e.g., stacking), the accuracy of the ML classifier improved following combining three widely used ML models (e.g., RFE-SVM, EmbeddingLR-LR, and RFE-AdaBoost). Decision curve analysis showed the merits of the ensemble classifiers, as the curves of the top 3 initial classifiers varied a lot in predicting JOA recovery rate in DCM patients. Conclusions: The ensemble classifiers successfully predict the JOA recovery rate in DCM patients, which showed a high potential for assisting physicians in managing DCM patients and making full use of medical resources. [ABSTRACT FROM AUTHOR]

Psychological distress in women represents a critical aspect; however, few studies have comprehensively addressed how health management strategies, consideration of future consequences (CFC), cognitive reappropriation, and self-efficacy may influence this phenomenon, especially in contexts that require a comprehensive approach to women's mental health. This study aimed to evaluate the effects of future consequence consideration, cognitive reappraisal, and self-efficacy in health management on psychological distress in American women, as well as to demonstrate their equivalence according to age. Data from 2788 women responding to the 2019 Health Information National Trends Survey 5, cycle 3 were used. A predictive model was considered using structural equation modeling (SEM). Cognitive reappraisal and self-efficacy in health management (SEHM) have a significant mediating effect between CFC and psychological distress. In addition, the invariance of the SEM model was presented as a function of age. The findings underscore the importance of incorporating cognitive components into mental health interventions for women, focusing on modifying thoughts and perceptions about the efficacy of health care to mitigate negative emotions. [ABSTRACT FROM AUTHOR]

Purpose: The current study used behavioral measures of discourse complexity and story recall accuracy in an expository discourse task to distinguish older adults testing within range of cognitive impairment according to a standardized cognitive screening tool in a sample of self-reported healthy older adults. Method: Seventy-three older adults who self-identified as healthy completed an expository discourse task and neuropsychological screener. Discourse data were used to classify participants testing within range of cognitive impairment using multiple machine learning algorithms and stability analysis for identifying reliably predictive features in an effort to maximize prediction accuracy. We hypothesized that a higher rate of pronoun use and lower scores on story recall would best classify older adults testing within range of cognitive impairment. Results: The highest classification accuracy exploited a single variable in a remarkably intuitive way: using 66% story recall as a cutoff for cognitive impairment. Forcing this decision tree model to use more features or increasing its complexity did not improve accuracy. Permutation testing confirmed that the 77% accuracy and 0.18 Brier skill score achieved by the model were statistically significant (p < .00001). Conclusions: These results suggest that expository discourse tasks that place demands on executive functions, such as working memory, can be used to identify aging adults who test within range of cognitive impairment. Accurate representation of story elements in working memory is critical for coherent discourse. Our simple yet highly accurate predictive model of expository discourse provides a promising assessment for easy identification of cognitive impairment in older adults. [ABSTRACT FROM AUTHOR]

This study investigated whether the use of peak force (PF) obtained from multiple joint positions during isometric squat (IsoSqt) can predict of 1-repetition maximum (1RM) squat with high precision. Twenty-nine male and female resistance-trained athletes (age = 23.0 ± 3.7 years, height = 1.57 ± 0.06 m, body mass = 56.6 ± 9.1 kg, 1RM squat = 104.6 ± 24.2) performed the 1RM squat, and IsoSqt at 60°, 90°, and 120° knee angles on 3 separate occasions. Peak force obtained from IsoSqt at all positions significantly correlated with 1RM squat (r = 0.684-0.940, p < 0.05). Linear regression analysis based on group data showed r² = 0.903, based on male data only, r² = 0.826, and based on female data only, r² = 0.855. Predicted 1RM squat using linear regression equations based on group, male-only, and female-only data showed nonsignificant differences of 0.02-0.6% from actual 1RM squat (p = 0.812-0.947, 95% CI = -8.00 to 10.08), with error of estimate of 1.5-2.3%. In addition, Bland-Altman analysis showed a mean bias of -1.04 to 0.35 kg, 95% CI = -10.08 to 8.00 kg. The results showed that using PF obtained from IsoSqt at multiple positions was able to predict 1RM squat with low difference from actual 1RM squat. Furthermore, using linear regression equation derived from gender-specific data was able to predict 1RM squat with higher precision than that based on group data. [ABSTRACT FROM AUTHOR]

Purpose: Target word activation in picture naming was explored in children who stutter (CWS) and typically fluent children (TFC) using event-related potentials (ERPs). Method: A total of 18 CWS and 16 TFC completed a task combining picture naming and probe word identification. On each trial, a picture-to-be-named was followed by an auditory probe word-to-be-identified; the probe was identical (Identity condition) or Unrelated to the picture name. ERPs were recorded from probe onset. Attenuation of the N400 ERP component was predicted to Identity versus Unrelated trials (N400 priming). Between-groups differences in amplitude, timing, and topography of N400 priming were explored. Results: Naming was more accurate on Identity versus Unrelated trials. Probe word identification accuracy was not affected by Condition. N400 priming was detected, indicating that self-generated picture names facilitated semantic processing of identical probes. This effect was larger in amplitude in CWS versus TFC. Unexpectedly, an N400-preceding, frontally maximal, positive-going ERP component--associated with expectancy processing--was larger in amplitude to Unrelated versus Identity trials. This effect was smaller in CWS versus TFC. Conclusions: A larger N400 priming effect in CWS versus TFC reflects a tendency toward more extensive semantic processing in picture naming in CWS. A smaller Condition effect on frontally maximal, positive-going, N400-preceding ERP activity in CWS versus TFC indicates a reduced ability to form expectancies about the lexical and/or phonological identity of probe words in CWS. Both effects may point to inefficient activation of target words in picture naming in CWS. [ABSTRACT FROM AUTHOR]

Simple Summary: Seizures are one of the most common and severe symptoms of meningioma, leading to increased morbidity and mortality in the affected patients. Therefore, seizure prevention represents an important goal in the treatment of meningioma patients. For this purpose, our study aims to identify predictors for the occurrence of preoperative and postoperative seizures in meningioma. Neurosurgical tumor resection was demonstrated as an effective treatment of seizures in meningioma patients but is also associated with a moderate risk of new-onset seizures after surgery. The present study identified several independent predictors for seizures in meningioma that could contribute to improved seizure treatment and a deeper understanding of the occurrence of seizures in meningioma patients. Seizure is a common symptom of meningioma that has a major impact on patients' quality of life. The purpose of this study was to identify predictive factors for the occurrence of preoperative and postoperative seizures. The data of patients with resection of histologically confirmed meningioma at University Hospital Leipzig from 2009 to 2018 were retrospectively examined. Univariate and multivariate logistic regression analyses of different factors influencing seizure outcome were performed. The male gender was identified as an independent positive predictor for preoperative seizures (odds ratio [OR] 1.917 [95% confidence interval {CI} 1.044–3.521], p = 0.036), whereas headache (OR 0.230 [95% CI 0.091–0.582], p = 0.002) and neurological deficits (OR 0.223; [95% CI 0.121–0.410], p < 0.001) were demonstrated to be negative predictive factors. Sensorimotor deficit after surgery (OR 4.490 [95% CI 2.231–9.037], p < 0.001) was found to be a positive predictor for the occurrence of postoperative seizures. The identified predictors for the occurrence of seizures in meningioma can contribute to improving seizure treatment and patients' quality of life. [ABSTRACT FROM AUTHOR]

Simple Summary: Preoperative risk prediction prior to oncologic esophagectomy is crucial for assisting surgeons in accurate patient selection and patients in their informed decision making. A new risk stratification tool, the IESG prediction model, was recently introduced, categorizing patients into different risk levelsMachine learning is a subfield of artificial intelligence and may allow for a more accurate identification of patients at risk. Therefore, we evaluated the IESG risk model and compared its performance with ML models. We found that the IESG risk model provided an overall adequate risk estimation. However, ML showed better results in the accurate risk stratification of patients, demonstrating its potential as a novel and powerful approach for future patient assessment. Risk prediction prior to oncologic esophagectomy is crucial for assisting surgeons and patients in their joint informed decision making. Recently, a new risk prediction model for 90-day mortality after esophagectomy using the International Esodata Study Group (IESG) database was proposed, allowing for the preoperative assignment of patients into different risk categories. However, given the non-linear dependencies between patient- and tumor-related risk factors contributing to cumulative surgical risk, machine learning (ML) may evolve as a novel and more integrated approach for mortality prediction. We evaluated the IESG risk model and compared its performance to ML models. Multiple classifiers were trained and validated on 552 patients from two independent centers undergoing oncologic esophagectomies. The discrimination performance of each model was assessed utilizing the area under the receiver operating characteristics curve (AUROC), the area under the precision–recall curve (AUPRC), and the Matthews correlation coefficient (MCC). The 90-day mortality rate was 5.8%. We found that IESG categorization allowed for adequate group-based risk prediction. However, ML models provided better discrimination performance, reaching superior AUROCs (0.64 [0.63–0.65] vs. 0.44 [0.32–0.56]), AUPRCs (0.25 [0.24–0.27] vs. 0.11 [0.05–0.21]), and MCCs (0.27 ([0.25–0.28] vs. 0.15 [0.03–0.27]). Conclusively, ML shows promising potential to identify patients at risk prior to surgery, surpassing conventional statistics. Still, larger datasets are needed to achieve higher discrimination performances for large-scale clinical implementation in the future. [ABSTRACT FROM AUTHOR]

Simple Summary: Prostate mpMRI is currently the most widely used image diagnosis approach to detect prostate cancer, while the PI-RADS system was developed to standardize and improve the accuracy of suspicious lesion identification on MRI. However, there still remain several limitations including inter-individual inconsistencies and naked-eye insufficiency. This study aims to apply AI technology to image interpretation to enhance diagnostic efficiency and explore the use of T2-weighted image-based stimulated biopsy in predicting prostate cancer (PCa). Using 820 lesions from The Cancer Imaging Archive database and 83 lesions from Hong Kong Queen Mary Hospital, we constructed 18 machine-learning models based on three algorithms and conducted both internal and external validation. We found that the logistic regression-based model provides additional diagnostic value to the PI-RADS in predicting PCa. Background: Currently, prostate cancer (PCa) prebiopsy medical image diagnosis mainly relies on mpMRI and PI-RADS scores. However, PI-RADS has its limitations, such as inter- and intra-radiologist variability and the potential for imperceptible features. The primary objective of this study is to evaluate the effectiveness of a machine learning model based on radiomics analysis of MRI T2-weighted (T2w) images for predicting PCa in prebiopsy cases. Method: A retrospective analysis was conducted using 820 lesions (363 cases, 457 controls) from The Cancer Imaging Archive (TCIA) Database for model development and validation. An additional 83 lesions (30 cases, 53 controls) from Hong Kong Queen Mary Hospital were used for independent external validation. The MRI T2w images were preprocessed, and radiomic features were extracted. Feature selection was performed using Cross Validation Least Angle Regression (CV-LARS). Using three different machine learning algorithms, a total of 18 prediction models and 3 shape control models were developed. The performance of the models, including the area under the curve (AUC) and diagnostic values such as sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV), were compared to the PI-RADS scoring system for both internal and external validation. Results: All the models showed significant differences compared to the shape control model (all p < 0.001, except SVM model PI-RADS+2 Features p = 0.004, SVM model PI-RADS+3 Features p = 0.002). In internal validation, the best model, based on the LR algorithm, incorporated 3 radiomic features (AUC = 0.838, sensitivity = 76.85%, specificity = 77.36%). In external validation, the LR (3 features) model outperformed PI-RADS in predictive value with AUC 0.870 vs. 0.658, sensitivity 56.67% vs. 46.67%, specificity 92.45% vs. 84.91%, PPV 80.95% vs. 63.64%, and NPV 79.03% vs. 73.77%. Conclusions: The machine learning model based on radiomics analysis of MRI T2w images, along with simulated biopsy, provides additional diagnostic value to the PI-RADS scoring system in predicting PCa. [ABSTRACT FROM AUTHOR]

Background: Identifying the minimal clinically important difference (MCID) contributes to the ability to determine the efficacy of physiotherapy interventions and make good clinical decisions. Purpose: The purpose of this study was to estimate the MCID for 6-minute walking distance (6MWD) among inpatients with subacute cardiac disease using multiple anchor-based methods. Methods: This study was a secondary data analysis using only data from a multicenter longitudinal observational study in which 6MWD was measured at two time points. Based on the changes in 6MWD between baseline measurement and follow-up approximately 1 week after baseline measurement, the global rating of change scales (GRCs) of patients and physiotherapists, anchor method receiver operator operating characteristic curves, predictive models, and adjusted models were used to calculate the MCID. Results: Participants comprised 35 patients. Mean (standard deviation) 6MWD was 228.9 m (121.1 m) at baseline and 270.1 m (125.0 m) at follow-up. MCID for each GRC was 27.5–35.6 m for patients and 32.5–38.6 m for physiotherapists. Conclusion: The MCID in 6MWD in patients with subacute cardiovascular disease is 27.5–38.6 m. This value may be useful in determining the effectiveness of physiotherapy interventions and for decision-making. [ABSTRACT FROM AUTHOR]

Simple Summary: This study introduces a deep learning model designed to predict the 5-year recurrence risk of lung adenocarcinoma based on histopathology images. Using a dataset of 189 patients with 455 histopathology slides, our model demonstrated superior performance in risk stratification, achieving a hazard ratio of 2.29 (95% CI: 1.69–3.09, p < 0.005). This outperforms several existing deep learning methods, showcasing the potential of deep learning in automatically predicting lung adenocarcinoma recurrence risk. The superior performance of this model underscores the potential for deep learning models to be integrated into clinical workflows for more accurate and automated risk assessment in lung adenocarcinoma. This could lead to more personalized treatment strategies and better patient outcomes. Lung cancer is the leading cause of cancer-related death in the United States. Lung adenocarcinoma (LUAD) is one of the most common subtypes of lung cancer that can be treated with resection. While resection can be curative, there is a significant risk of recurrence, which necessitates close monitoring and additional treatment planning. Traditionally, microscopic evaluation of tumor grading in resected specimens is a standard pathologic practice that informs subsequent therapy and patient management. However, this approach is labor-intensive and subject to inter-observer variability. To address the challenge of accurately predicting recurrence, we propose a deep learning-based model to predict the 5-year recurrence of LUAD in patients following surgical resection. In our model, we introduce an innovative dual-attention architecture that significantly enhances computational efficiency. Our model demonstrates excellent performance in recurrent risk stratification, achieving a hazard ratio of 2.29 (95% CI: 1.69–3.09, p < 0.005), which outperforms several existing deep learning methods. This study contributes to ongoing efforts to use deep learning models for automatically learning histologic patterns from whole slide images (WSIs) and predicting LUAD recurrence risk, thereby improving the accuracy and efficiency of treatment decision making. [ABSTRACT FROM AUTHOR]

INTRODUCTION: A prolonged operative time of lithotripsy with ureteroscopy for urolithiasis increases the risk of infectious complications; however, few reports have investigated the factors prolonging the operative time for ureteral stones. We investigated the factors associated with longer operative time in ureteroscopy for ureteral stones. METHODS: This retrospective cohort study analyzed patients who underwent retrograde ureteroscopic lithotripsy for ureteral stones and achieved an endoscopic stone-free status between April 2019 and July 2022. Patients were classified into two groups based on an operative time of ≥90 minutes or <90 minutes. We compared the patient and stone characteristics and surgical outcomes, and investigated the factors associated with a prolonged operative time. RESULTS: The cohort comprised 519 patients, with 58 patients in the group with an operative time of ≥90 minutes. Compared to the shorter operative time group, the longer operative time group had a significantly greater proportion of males, stone diameter, stone volume, and Hounsfield units of stone; additionally, the longer operative time group had higher prevalences of endoscopic findings of edema, polyps, and mucosa-stone adherence. Multivariable analysis showed that stone size >10 mm (odds ratio 4.05), polyps (odds ratio 2.40), and mucosal adherence (odds ratio 3.51) were significantly associated with an operative time exceeding 90 minutes. There were no significant differences between the two groups in the incidences of postoperative fever and systemic inflammatory response syndrome. CONCLUSIONS: Stone size, endoscopic findings of polyps, and mucosa-stone adherence were independent factors associated with a longer operative time. [ABSTRACT FROM AUTHOR]

Objective To develop a machine learning-based prediction model for identifying hyperuricemic participants at risk of developing gout. Methods A retrospective nationwide Israeli cohort study used the Clalit Health Insurance database of 473 124 individuals to identify adults 18 years or older with at least two serum urate measurements exceeding 6.8 mg/dl between January 2007 and December 2022. Patients with a prior gout diagnosis or on gout medications were excluded. Patients' demographic characteristics, community and hospital diagnoses, routine medication prescriptions and laboratory results were used to train a risk prediction model. A machine learning model, XGBoost, was developed to predict the risk of gout. Feature selection methods were used to identify relevant variables. The model's performance was evaluated using the receiver operating characteristic area under the curve (ROC AUC) and precision-recall AUC. The primary outcome was the diagnosis of gout among hyperuricemic patients. Results Among the 301 385 participants with hyperuricemia included in the analysis, 15 055 (5%) were diagnosed with gout. The XGBoost model had a ROC-AUC of 0.781 (95% CI 0.78–0.784) and precision-recall AUC of 0.208 (95% CI 0.195–0.22). The most significant variables associated with gout diagnosis were serum uric acid levels, age, hyperlipidemia, non-steroidal anti-inflammatory drugs and diuretic purchases. A compact model using only these five variables yielded a ROC-AUC of 0.714 (95% CI 0.706–0.723) and a negative predictive value (NPV) of 95%. Conclusions The findings of this cohort study suggest that a machine learning-based prediction model had relatively good performance and high NPV for identifying hyperuricemic participants at risk of developing gout. [ABSTRACT FROM AUTHOR]

Objectives: To explore socio-behavioral, clinical, and imaging findings associated with antepartum intimate partner violence (IPV) and aid in risk stratification of at-risk individuals. Methods: We analyzed electronic medical records during indexed pregnancies for 108 pregnant patients who self-reported antepartum IPV (cases) and 106 age-matched pregnant patients who did not self-report antepartum IPV (controls). Sociodemographic, clinical, and radiology data were analyzed via chi-squared and Fisher's exact tests with p < 0.05 as the threshold for significance. Stepwise logistic regression was applied to derive a risk prediction model. Results: The proportion of cases reporting emotional IPV (76% vs. 52%) and/or physical IPV (45% vs. 31%) during pregnancy significantly increased from prior to pregnancy. Cases were significantly more likely to report prepregnancy substance use (odds ratio [OR] = 2.60; 95% confidence interval [CI]: 1.13–5.98), sexually transmitted infections (OR = 3.48; 95%CI: 1.64–7.37), abortion (OR = 3.17; 95%CI: 1.79, 5.59), and preterm birth (OR = 5.97; 95%CI: 1.69–21.15). During pregnancy, cases were more likely to report unstable housing (OR = 5.26; 95%CI: 2.67–10.36), multigravidity (OR = 2.83; 95%CI: 1.44–5.58), multiparity (OR = 3.75; 95%CI: 1.72–8.20), anxiety (OR = 3.35; 95%CI: 1.85–6.08), depression (OR = 5.58; 95%CI: 3.07–10.16), substance use (OR = 2.92; 95%CI: 1.28–6.65), urinary tract infection (UTI) (OR = 3.26; 95%CI: 1.14–9.32), intrauterine growth restriction (OR = 10.71; 95%CI: 1.35–85.25), and cesarean delivery (OR = 2.25; 95%CI: 1.26–4.02). Cases had significantly more OBGYN abnormalities on imaging and canceled more radiological studies (OR = 5.31). Logistic regression found housing status, sexually transmitted infection history, preterm delivery history, abortion history, depression, and antepartum UTI predictive of antepartum IPV. The risk prediction model achieved good calibration with an area under the curve of 0.79. Conclusions: This study identifies significant disparities among patients experiencing antepartum IPV, and our proposed risk prediction model can inform risk assessment in this setting. [ABSTRACT FROM AUTHOR]

Objective: To compare the Guy’s, S.T.O.N.E, and CROES nomograms for predicting stone-free status in patients who underwent percutaneous nephrolithotomy for renal stones. Materials and Methods: The data of 1114 patients who underwent percutaneous nephrolithotomy for renal calculi between 11/2008 and 08/2018 in our clinic were retrospectively reviewed. Various parameters evaluated by preoperative computed tomography and the scoring systems of the patients and postoperative stone-free status were compared. Results: Out of 1000 patients who met the study criteria. Gender, body mass index, and stone density were not statistically different between the group with residual stones and the stone-free group. However, stone size, number of renal accesses, duration of fluoroscopy usage, duration of operation, number of stones, and complication rate were significantly higher in the group with residual stones than in the stone-free group. A statistically significant correlation was found between the postoperative stone-free rate and scoring systems. The applicability and preoperative prediction ability of all three systems were evaluated by receiver operating characteristic analysis. The area under the curve (AUC) was detected in the Guy’s, CROES, and S.T.O.N.E scoring system (AUC: 0.642, 0.665, 0.592 respectively). Conclusion: In this study, where the perioperative and postoperative results of 1000 patients were evaluated, we found that all three scoring systems could predict the stone-free rate. “We believe that the use of these scoring systems before surgery can guide surgeons.” [ABSTRACT FROM AUTHOR]

The six‐minute walking test (6MWT) is commonly used to measure functional capacity in field settings, primarily through the distance covered. This study aims to establish reference curves for the six‐minute walking distance (6MWD) and peak heart rate (PHR) and develop a predictive equation for cardiovascular capacity in Tunisian children and adolescents. A total of 1501 participants (706 boys and 795 girls), aged 10–18 years, were recruited from schools in Tunisia. The Lambda (L), Mu (M), and Sigma (S) methods (LMS method) were employed to develop smoothed percentile curves for 6MWD and PHR. Multivariate linear regression was utilized to formulate a prediction equation for 6MWD. Smoothed percentiles (3rd, 10th, 25th, 50th, 75th, 90th, and 97th) for 6MWD and PHR were presented with age. All variables showed a strong positive correlation (p < 0.001) with a six‐minute walking distance (r ranged from 0.227 to 0.558 for girls and from 0.309 to 0.610 for boys), except resting heart rate, which showed a strong negative correlation (girls: r = −0.136; boys: r = −0.201; p < 0.001). Additionally, PHR showed a weak correlation (p > 0.05). The prediction equations, based on age as the primary variable, were established for both genders. For boys: 6MWD = 66.181 + 38.142 × Age (years) (R2 = 0.372; Standard Error of Estimate (SEE) = 122.13), and for girls: 6MWD = 105.535 + 28.390 × Age (years) (R2 = 0.312; SEE = 103.66). The study provides normative values and predictive equations for 6MWD and PHR in Tunisian children and adolescents. These findings offer essential tools for identifying, monitoring, and interpreting cardiovascular functional deficits in clinical and research settings. Highlights: Established reference curves for six‐minute walking distance (6MWD) and peak heart rate (PHR) from the data of 1501 children and adolescents aged 10–18 years were recruited from multicenter schools in Tunisia.Developed predictive equations for 6MWD based on age using multivariate linear regression.Provides normative values and predictive equations for 6MWD and PHR, offering valuable tools for identifying and monitoring cardiovascular functional deficits in clinical and research settings. [ABSTRACT FROM AUTHOR]

Polysocial risk scores were recently proposed as a strategy for improving the clinical relevance of knowledge about social determinants of health. Our objective in this study was to assess whether the polysocial risk score model improves prediction of cognition and all-cause mortality in middle-aged and older adults beyond simpler models including a smaller set of key social determinants of health. We used a sample of 13 773 individuals aged ≥50 years at baseline from the 2006-2018 waves of the Health and Retirement Study, a US population-based longitudinal cohort study. Four linear mixed models were compared: 2 simple models including a priori–selected covariates and 2 polysocial risk score models which used least absolute shrinkage and selection operator (LASSO) regularization to select covariates among 9 or 21 candidate social predictors. All models included age. Predictive accuracy was assessed via R 2 and root mean-squared prediction error (RMSPE) using training/test split validation and cross-validation. For predicting cognition, the simple model including age, race, sex, and education had an R 2 value of 0.31 and an RMSPE of 0.880. Compared with this, the most complex polysocial risk score selected 12 predictors (R 2 = 0.35 and RMSPE = 0.858; 2.2% improvement). For all-cause mortality, the simple model including age, race, sex, and education had an area under the receiver operating characteristic curve (AUROC) of 0.747, while the most complex polysocial risk score did not demonstrate improved performance (AUROC = 0.745). Models built on a smaller set of key social determinants performed comparably to models built on a more complex set of social "risk factors." [ABSTRACT FROM AUTHOR]

The objective of this research was to compare social, emotional, and cognitive determinants of depression and anxiety symptoms and mental wellbeing using the two-continua model of mental health and illness in two age groups. The two-continua model proposes that mental health and mental illness are not two ends of the same spectrum but are, in fact, two separate but related spectrums. This cross-sectional study used a multi-age group comparison approach to identify the relative predictive strengths of risk and protective factors for the two-continua model of mental health and mental illness. Participants (n = 458: 251 16–25 year olds, and 207 35–64 year olds) completed an online survey that measured the outcome variables of depression and anxiety symptoms and mental wellbeing. Independent variables from three conceptual areas in psychology—(1) positive psychology: optimism, pessimism, and accomplishment; (2) emotion regulation: cognitive reappraisal and expressive suppression; and (3) interpersonal theories: belonging and relationship with parent/child—were measured to ascertain the determinants of these outcome variables. The all-variables models explained 58–68% of the variance in depression symptoms, 77–80% in mental wellbeing, and 26–43% in anxiety symptoms. For both groups, the strongest predictor of mental wellbeing in these models was accomplishment. The strongest predictors in these models of mental illness symptoms differed between groups: belonging in the younger group and accomplishment in the older group. Programs targeting belonging and accomplishment could be highly effective in promoting mental wellbeing and reducing mental ill-health for these groups. Interventions require contextual investigation to locate drivers of mental wellbeing and illness for different age groups prior to implementation. [ABSTRACT FROM AUTHOR]

Purpose: Cardiorespiratory fitness (CRF) is a criticalmarker of overall health and a key predictor of morbidity andmortality, but the existing prediction equations for CRF are primarily derived from general populations and may not be suitable for patients with obesity. Methods: Predicted CRF from different non-exercise prediction equations was compared with measured CRF of patients with obesity who underwent maximal cardiopulmonary exercise testing (CPET). Multiple linear regression was used to develop a population-specific nonexercise CRF prediction model for treadmill exercise including age, sex, weight, height, and physical activity level as determinants. Results: Six hundred sixty patients underwent CPET during the study period. Within the entire cohort, R² values had a range of 0.24 to 0.46. Predicted CRF was statistically different from measured CRF for 19 of the 21 included equations. Only 50% of patients were correctly classified into the measured CRF categories according to predicted CRF. A multiple model for CRF prediction (mL⋅min-1) was generated (R² = 0.78) and validated using two crossvalidation methods. Conclusions: Most used equations provide inaccurate estimates of CRF in patients with obesity, particularly in cases of severe obesity and low CRF. Therefore, a new prediction equation was developed and validated specifically for patients with obesity, offering a more precise tool for clinical CPET interpretation and risk stratification in this population. [ABSTRACT FROM AUTHOR]

Introduction: The skill domains of fundamental movement skills (FMS), emphasizing gross motor movements, including locomotor skills (LMS) and object control skills (OCS), along with process- and product-oriented measures of FMS, may predict cardiorespiratory fitness (CRF) and muscular fitness (MF) differently. This study investigates the longitudinal relationship from early childhood FMS, focusing on process-oriented LMS and OCS and product-oriented FMS, to CRF and MF in late childhood. Methods: The study involved 441 Finnish children (49.9% female; mean age at baseline, 5.5 yr) over a 6-yr period. FMS was evaluated using the Test of Gross Motor Development, third version, for process-oriented LMS and OCS, and the Körperkoordinationstest Für Kinder (KTK) was used to evaluate the product-oriented FMS. CRF was assessed through the total number of laps completed in the 20-m shuttle run test, whereas MF was measured via repetitions of curl-ups and push-ups. Employing a two-level cross-classified regression analysis and Cholesky decomposition, this study aimed to determine the contributions of product-oriented KTK and process-oriented LMS and OCS. Adjustments for variations in age, measurement intervals, and maturation were achieved through residualization. In addition, gender and body mass index were incorporated as covariates in the analysis. Results: The analysis revealed that process-oriented LMS (CRF: ΔR² = 0.016; MF: ΔR² = 0.014) significantly predicted later health-related fitness, whereas OCS did not. However, KTK exhibited a better ability to predict both CRF (ΔR² = 0.092) and MF (ΔR² = 0.032), overshadowing process-oriented measures. Conclusions: In conclusion, the findings suggest that KTK, which potentially encompasses a broader spectrum of fitness elements along with FMS, more effectively predicts health-related fitness components than process-oriented FMS. [ABSTRACT FROM AUTHOR]

Purpose: Habitual strength and power-demanding activities of daily life may support the maintenance of adequate lower extremity functioning with aging, but this has been sparingly explored. Hence, we examined whether the characteristics of free-living sit-to-stand (STS) transitions predict a decline in lower extremity functioning over a 4-yr follow-up. Methods: A total of 340 community-dwelling older adults (60% women; age 75, 80, or 85 yr) participated in this prospective cohort study. At baseline, a thigh-worn accelerometer was used continuously (3-7 d) to monitor the number and intensity of free-living STS transitions. A decline in lower extremity functioning was defined as a drop of ≥2 points in the Short Physical Performance Battery (SPPB) from baseline to follow-up. Maximal isometric knee extension strength was measured in the laboratory. Results: Eighty-five participants (75%women) declined in SPPB over 4 yr. After adjusting for age, sex, and baseline SPPB points, higher free-living peak STS angular velocity (odds ratio (OR), 0.70; 95% confidence interval (CI), 0.52-0.92, per 20°⋅s-1 increase) protected against a future decline. When adjusting the model for maximal isometric knee extension strength, the statistical significance was attenuated (OR = 0.72; 95% CI = 0.54-0.96, per 20°⋅s-1 increase). Conclusions: Performing STS transitions at higher velocities in the free-living environment can prevent a future decline in lower extremity function. This indicates that changes in daily STS behavior may be useful in the early identification of functional loss. Free-living peak STS angular velocity may be a factor underlying the longitudinal association of lower extremity strength and performance. [ABSTRACT FROM AUTHOR]

The aims of the study were to build models using logistic regression analysis of flexibility and strength tests to prospectively predict risk factors for anterior cruciate ligament tear (ACL-tear) in female soccer (FS) players, and to determine training cut-off for risk factors of the predictive model built. A prospective cohort study of 95 female players (aged 14–33 years) was conducted. Age, anthropometric data, soccer history, lower limb range of motion (ROM) and hip maximal isometric strength (MIS) were measured. At the prospective follow-up after 12 months, 7.4% of the players had developed an ACL-tear. The model showed a significant relationship (χ2(93) = 30.531, p < 0.001) between the ACL-tear and the predictor variables (leg length, HAD-NH [hip adduction] MIS, asymmetric ROM [ankle dorsiflexion with knee extended (AD-KE) and with knee flexed (AD-KF), and HE (hip extension)], hip ROM [HIR (internal rotation) and HAB (abduction)]). The Akaike Information Criteria (AIC) and Bayesian Information Criteria (BIC) for model fit were 30.24 and 51.79, respectively. The value R2 showed good model fit, 76.5% for Nagelkerke´s R2, 71.4% for McFadden´s R2 and 67.5% for Tjur´s R2. For the screening test, cut-off for leg length of ≥0.40 m, for HIR ROM of ≤44º and for asymmetry of HE ROM of ≥5° were set, which have an acceptable (AUC ≥ 0.755) discriminatory ability for the development of ACL-tear. [ABSTRACT FROM AUTHOR]

Objective: Accurately predicting knee osteoarthritis (KOA) is essential for early detection and personalized treatment. We aimed to develop and test a magnetic resonance imaging (MRI)‐based joint space (JS) radiomic model (RM) to predict radiographic KOA incidence through neural networks by integrating meniscus and femorotibial cartilage radiomic features. Methods: In the Osteoarthritis Initiative cohort, participants with knees without radiographic KOA at baseline but at high risk for radiographic KOA were included. Patients' knees developed radiographic KOA, whereas control knees did not over four years. We randomly split the participants into development and test cohorts (8:2) and extracted features from baseline three‐dimensional double‐echo steady‐state sequence MRI. Model performance was evaluated using an area under the receiver operating characteristic curve (AUC), sensitivity, and specificity in both cohorts. Nine resident surgeons performed the reader experiment without/with the JS‐RM aid. Results: Our study included 549 knees in the development cohort (275 knees of patients with KOA vs 274 knees of controls) and 137 knees in the test cohort (68 knees of patients with KOA vs 69 knees of controls). In the test cohort, JS‐RM had a favorable accuracy for predicting the radiographic KOA incidence with an AUC of 0.931 (95% confidence interval [CI] 0.876–0.963), a sensitivity of 84.4% (95% CI 83.9%–84.9%), and a specificity of 85.6% (95% CI 85.2%–86.0%). The mean specificity and sensitivity of resident surgeons through MRI reading in predicting radiographic KOA incidence were increased from 0.474 (95% CI 0.333–0.614) and 0.586 (95% CI 0.429–0.743) without the assistance of JS‐RM to 0.874 (95% CI 0.847–0.901) and 0.812 (95% CI 0.742–0.881) with JS‐RM assistance, respectively (P < 0.001). Conclusion: JS‐RM integrating the features of the meniscus and cartilage showed improved predictive values in radiographic KOA incidence. [ABSTRACT FROM AUTHOR]