Search

Background: This multicentre retrospective observational study aims to develop imaging-based prognostic and predictive models for relapsed/refractory (R/R) B-cell lymphoma patients undergoing CAR-T therapy by integrating clinical data and imaging features. Specifically, our aim was to predict 3- and 6-month treatment response, overall survival (OS), progression-free survival (PFS), and the occurrence of the immune effector cell-associated neurotoxicity syndrome (ICANS). Results: Sixty-five patients of R/R B-cell lymphoma treated with CAR-T cells in two centres were included. Pre-infusion [18F]FDG PET/CT scans and clinical data were systematically collected, and imaging features, including kurtosis, entropy, maximum diameter, standardized uptake value (SUV) related features (SUVmax, SUVmean, SUVstd, SUVmedian, SUVp25, SUVp75), total metabolic tumour volume (MTVtotal), and total lesion glycolysis (TLGtotal), were extracted using the Quibim platform. The median age was 62 (range 21–76) years and the median follow-up for survivors was 10.47 (range 0.20–45.80) months. A logistic regression model accurately predicted neurotoxicity (AUC: 0.830), and Cox proportional-hazards models for CAR-T response at 3 and 6 months demonstrated high accuracy (AUC: 0.754 and 0.818, respectively). Median predicted OS after CAR-T therapy was 4.73 months for high MTVtotal and 37.55 months for low MTVtotal. Median predicted PFS was 2.73 months for high MTVtotal and 11.83 months for low MTVtotal. For all outcomes, predictive models, combining imaging features and clinical variables, showed improved accuracy compared to models using only clinical variables or imaging features alone. Conclusion: This study successfully integrates imaging features and clinical variables to predict outcomes in R/R B-cell lymphoma patients undergoing CAR-T. Notably, the identified MTVtotal cut-off effectively stratifies patients, as evidenced by significant differences in OS and PFS. Additionally, the predictive models for neurotoxicity and CAR-T response show promising accuracy. This comprehensive approach holds promise for risk stratification and personalized treatment strategies which may become a helpful tool for optimizing CAR-T outcomes in R/R lymphoma patients. [ABSTRACT FROM AUTHOR]

Background: The rate of 30-day all-cause hospital readmissions can affect the funding a hospital receives. An accurate and reliable readmission prediction model could save money and increase quality-of-care. Few projects have explored formulating this task as a survival prediction problem, where models can exploit a real-valued time-to-readmission target. This paper demonstrates the effectiveness of a survival-inspired readmission model, especially when paired with a longitudinal patient representation that is agnostic to disease-cohort and predictive task. Methods: We forecast readmissions for a population-level cohort of 421,088 patients discharged in 2015 and 2016 from hospitals in Alberta, Canada. Clinical features and sequences of historical medical codes (calculated from at least four full years prior to discharge) from linked administrative sources serve as model inputs. We trained binary 30-day readmission models (XGBoost and a Deep Neural Network) and time-to-event readmission models (CoxPH and N-MTLR) with and without machine-learned medical knowledge at initialization, then compared against the popular LACE-based model using the AUROC score at 30 days (AUROC@30). Survival models are additionally evaluated using concordance, Integrated Brier, and L1-loss scores. Results: All models that utilize sequence features markedly out-perform even the best models trained on only clinical features. Further, a time-to-event target improves predictive performance at 30 days, given the same model inputs and architecture. N-MTLR, using solely sequence inputs and initialized with pre-learned medical knowledge, achieves an average AUROC@30 of 0.8460 over five folds with a standard deviation of 0.003. All trained models match or out-perform the LACE baseline of 0.6587±0.003. Conclusion: Sequences of administrative medical codes contain rich predictive information for forecasting readmissions, and embedding medical knowledge a priori using machine learning provides readmission models an advantageous foundation for training. When combined with a model that can leverage a time-to-event target, excellent performance is possible on the 30-day all-cause readmission task using only administrative data. [ABSTRACT FROM AUTHOR]

Accurate prediction of gastric cancer survival state is one of great significant tasks for clinical decision‐making. Many advanced machine learning classification techniques have been applied to predict the survival status of cancer patients in three or 5 years, however, many of them have a low sensitivity because of class imbalance. This is a non‐negligible problem due to the poor prognosis of gastric cancer patients. Furthermore, models in the medical domain require strong interpretability to increase their applicability. Due to the better performance and interpretability of the XGBoost model, we design a loss function taking into account cost sensitive and focal loss from the algorithm level for XGBoost to deal with the imbalance problem. We apply the improved model into the prediction of the survival status of gastric cancer patients and analyse the important related features. We use two types of indicators to evaluate the model, and we also design the confusion matrix of two models' predictive results to compare two models. The results show that the improved model has better performance. Furthermore, we calculate the importance of features related to survival with three different time periods and analyse their evolution, which are consistent with existing clinical research or further expand their research conclusions. These all support for clinically relevant decision‐making and has the potential to expand into survival prediction of other cancer patients. [ABSTRACT FROM AUTHOR]

Predicting survival outcomes in critical accidents has been a focal point in machine learning research. This study addresses several limitations of existing methods, including insufficient management of data imbalance, lack of emphasis on hyperparameter tuning, and proneness to overfitting. Many existing models struggle to generalize effectively on imbalanced datasets or depend on default hyperparameter settings, resulting in biased predictions. By integrating Principal Component Analysis (PCA), hyperparameter optimization, and resampling methods, as well as combining Edited Nearest Neighbors (ENN) with the Synthetic Minority Oversampling Technique (SMOTE), the model significantly improves predictive accuracy and model generalization. An ensemble model combining seven machine learning algorithms—Logistic Regression, Support Vector Machine, KNN, Random Forest, XGBoost, LightGBM, and CatBoost—was applied to predict survival outcomes. Stochastic Weighted Averaging (SWA) was applied to mitigate overfitting and enhance generalization. The accuracy increased from 91.97% to 94.89% after SWA was applied in this specific scenario. The combination of PCA-based dimensionality reduction, hyperparameter tuning, and resampling techniques (ENN + SMOTE) ensured the model handled data imbalance and optimized predictive accuracy. The final model demonstrated excellent performance, with Area Under the Curve (AUC) and Average Precision (AP) values both reaching 0.98, indicating high accuracy and precision. These improvements were validated using the Titanic dataset in a binary classification problem of predicting passenger survival. The results emphasize that ensemble learning, enhanced by SWA, offers a powerful framework for handling imbalanced and complex datasets, providing significant advancements in predictive modeling accuracy. This study provides insights into how machine learning techniques can be effectively combined to solve classification challenges in real-world scenarios. [ABSTRACT FROM AUTHOR]

Cancer survival estimation is essential for post-cancer patient care, cancer management policy building, and the development of tailored treatment plans. Existing survival estimation methods use censored data; therefore, standard machine learning methods can not be used directly. Some censoring-based semi-machine learning methods have recently been proposed; however, these methods pose challenges. They are less patient-specific and non-linear. Furthermore, they rely on single-modality features. These drawbacks result in lower survival estimation performance. To address these issues, this work proposes a framework named Deep-SEA. Compared to the state-of-the-art, Deep-SEA uses multi-modality features, i.e., clinical, radiology, and histology features. These features are analyzed with statistical methods, and only significant features are selected. Then, the baseline hazard of the Cox model is estimated using Breslow's estimator, which is optimized using stochastic gradient descent. Finally, the risk function, i.e., the parameters of our model, are estimated via an ANN with time as additional input. ANN makes it non-linear while training on the patient-specific features makes it more patient-specific than the state-of-the-art. We train and evaluate Deep-SEA on five datasets, including head, neck, and colorectal-liver cancer. We have achieved a Concordance-index (C-index) score of 0.7181, the highest compared to the state-of-the-art. Results and ablation studies on Deep-SEA suggest that the proposed method improves cancer survival estimation and can be applied to other estimations, such as cancer recurrence estimation. [ABSTRACT FROM AUTHOR]

Objective: Nasopharyngeal carcinoma (NPC) patients frequently present with distant metastasis (DM), which is typically associated with poor prognosis. This study aims to develop and apply machine learning models to predict DM, overall survival (OS), and cancer-specific survival (CSS) in NPC patients to provide optimal tools for improved predictive accuracy and performance. Methods: We retrieved over 8,000 NPC patient samples with associated clinical information from the Surveillance, Epidemiology, and End Results (SEER) database. Utilizing two methods for handling missing values--imputation or deletion--we created various cohorts: DM-all, DM-slim, OS-all, OS-slim, CSSall, and CSS-slim. Five machine learning models were deployed for the binary classification task of DM, and their performance was evaluated using the area under the curve (AUC). For the survival prediction tasks of OS and CSS, we constructed 45 combinations using nine survival machine learning algorithms. The Concordance Index (C-index), 5-year AUC, and Brier score assessed model accuracy. Patients were stratified into two risk groups for survival analysis, and the survival curves were presented. Results: This study examines the relationships between clinical factors and survival in NPC patients. The analysis, visualized through forest plots, indicates that demographic and clinical variables like gender, marital status, tumor grade, and stage significantly affect metastatic risks and survival. Specifically, factors such as advanced stages increase metastasis and survival risks, while enhanced treatments improve survival rates. In the cohort for DM prediction, results revealed that the random forest model was the most effective, with an AUC of 0.687. In contrast, when predicting overall survival (OS), the random survival forest (RSF) model consistently showed superior performance with the highest mean C-index of 0.802, a 5-year AUC of 0.857, and a Brier score of 0.167. Similarly, for cancer-specific survival (CSS) prediction, the RSF model demonstrated a mean C-index of 0.822, a 5-year AUC of 0.884, and a Brier score of 0.165. An online Shiny server was developed to allow the models to be used freely and efficiently via http://npcml.shinyapps.io/NPCpre. Conclusion: This study successfully established an online tool by machine learning models for NPC metastasis and survival prediction, providing valuable references for clinicians. [ABSTRACT FROM AUTHOR]

Objective: Coronavirus disease 2019 (COVID-19) can cause intubation and ventilatory support due to respiratory failure, and extubation failure increases mortality risk. This study, therefore, aimed to explore the feasibility of using specific biochemical and ventilator parameters to predict survival status among COVID-19 patients by using machine learning. Methods: This study included COVID-19 patients from Taipei Medical University-affiliated hospitals from May 2021 to May 2022. Sequential data on specific biochemical and ventilator parameters from days 0–2, 3–5, and 6–7 were analyzed to explore differences between the surviving (successfully weaned off the ventilator) and non-surviving groups. These data were further used to establish separate survival prediction models using random forest (RF). Results: The surviving group exhibited significantly lower mean C-reactive protein (CRP) levels and mean potential of hydrogen ions levels (pH) levels on days 0–2 compared to the non-surviving group (CRP: non-surviving group: 13.16 ± 5.15 ng/mL, surviving group: 10.23 ± 5.15 ng/mL; pH: non-surviving group: 7.32 ± 0.07, survival group: 7.37 ± 0.07). Regarding the survival prediction performanace, the RF model trained solely with data from days 0–2 outperformed models trained with data from days 3–5 and 6–7. Subsequently, CRP, the partial pressure of carbon dioxide in arterial blood (PaCO2), pH, and the arterial oxygen partial pressure to fractional inspired oxygen (P/F) ratio served as primary indicators in survival prediction in the day 0–2 model. Conclusions: The present developed models confirmed that early biochemical and ventilatory parameters—specifically, CRP levels, pH, PaCO2, and P/F ratio—were key predictors of survival for COVID-19 patients. Assessed during the initial two days, these indicators effectively predicted the likelihood of successful weaning of from ventilators, emphasizing their importance in early management and improved outcomes in COVID-19-related respiratory failure. [ABSTRACT FROM AUTHOR]

Ovarian cancer is a formidable health challenge that demands accurate and timely survival predictions to guide clinical interventions. Existing methods, while commendable, suffer from limitations in harnessing the temporal evolution of patient data and capturing intricate interdependencies among different data elements. In this paper, we present a novel methodology which combines Temporal Analysis and Graph Neural Networks (GNNs) to significantly enhance ovarian cancer survival rate predictions. The shortcomings of current processes originate from their disability to correctly seize the complex interactions amongst diverse scientific information units in addition to the dynamic modifications that arise in a affected person's nation over time. By combining temporal information evaluation and GNNs, our cautioned approach overcomes those drawbacks and, whilst as compared to preceding methods, yields a noteworthy 8.3% benefit in precision, 4.9% more accuracy, 5.5% more advantageous recall, and a considerable 2.9% reduction in prediction latency. Our method's Temporal Analysis factor uses longitudinal affected person information to perceive good-sized styles and tendencies that offer precious insights into the direction of ovarian cancer. Through the combination of GNNs, we offer a robust framework able to shoot complicated interactions among exclusive capabilities of scientific data, permitting the version to realize diffused dependencies that would affect survival results. Our paintings have tremendous implications for scientific practice. Prompt and correct estimation of the survival price of ovarian most cancers allows scientific experts to customize remedy regimens, manipulate assets efficiently, and provide individualized care to patients. Additionally, the interpretability of our version's predictions promotes a collaborative method for affected person care via way of means of strengthening agreement among scientific employees and the AI-driven selection help system. The proposed approach not only outperforms existing methods but also has the possible to develop ovarian cancer treatment by providing clinicians through a reliable tool for informed decision-making. Through a fusion of Temporal Analysis and Graph Neural Networks, we conduit the gap among data-driven insights and clinical practice, proposing a capable opportunity for refining patient outcomes in ovarian cancer management operations. [ABSTRACT FROM AUTHOR]

Background: The 3-variable number-of-risk-factors (NRF) model is a prognostic tool for patients undergoing palliative radiotherapy (PRT). However, there is little research on the NRF model for patients with painful non-bone-metastasis tumours treated with PRT, and the efficacy of the NRF model in predicting survival is unclear to date. Therefore, we aimed to assess the prognostic accuracy of a 3-variable NRF model in patients undergoing PRT for bone and non- bone-metastasis tumours. Methods: This was a secondary analysis of studies on PRT for bone-metastasis (BM) and PRT for miscellaneous painful tumours (MPTs), including non-BM tumours. Patients were grouped in the NRF model and survival was compared between groups. Discrimination was evaluated using a time-independent C-index and a time-dependent area under the receiver operating characteristic curve (AUROC). A calibration curve was used to assess the agreement between predicted and observed survival. Results: We analysed 485 patients in the BM group and 302 patients in the MPT group. The median survival times in the BM group for groups I, II, and III were 35.1, 10.1, and 3.3 months, respectively (P < 0.001), while in the MPT group, they were 22.1, 9.5, and 4.6 months, respectively (P < 0.001). The C-index was 0.689 in the BM group and 0.625 in the MPT group. In the BM group, time-dependent AUROCs over 2 to 24 months ranged from 0.738 to 0.765, while in the MPT group, they ranged from 0.650 to 0.689, with both groups showing consistent accuracy over time. The calibration curve showed a reasonable agreement between the predicted and observed survival. Conclusions: The NRF model predicted survival moderately well in both the BM and MPT groups. [ABSTRACT FROM AUTHOR]

Gliomas are aggressive brain tumors known for their heterogeneity, unclear borders, and diverse locations on Magnetic Resonance Imaging (MRI) scans. These factors present significant challenges for MRI-based segmentation, a crucial step for effective treatment planning and monitoring of glioma progression. This study proposes a novel deep learning framework, ResNet Multi-Head Attention U-Net (ResMHA-Net), to address these challenges and enhance glioma segmentation accuracy. ResMHA-Net leverages the strengths of both residual blocks from the ResNet architecture and multi-head attention mechanisms. This powerful combination empowers the network to prioritize informative regions within the 3D MRI data and capture long-range dependencies. By doing so, ResMHA-Net effectively segments intricate glioma sub-regions and reduces the impact of uncertain tumor boundaries. We rigorously trained and validated ResMHA-Net on the BraTS 2018, 2019, 2020 and 2021 datasets. Notably, ResMHA-Net achieved superior segmentation accuracy on the BraTS 2021 dataset compared to the previous years, demonstrating its remarkable adaptability and robustness across diverse datasets. Furthermore, we collected the predicted masks obtained from three datasets to enhance survival prediction, effectively augmenting the dataset size. Radiomic features were then extracted from these predicted masks and, along with clinical data, were used to train a novel ensemble learning-based machine learning model for survival prediction. This model employs a voting mechanism aggregating predictions from multiple models, leading to significant improvements over existing methods. This ensemble approach capitalizes on the strengths of various models, resulting in more accurate and reliable predictions for patient survival. Importantly, we achieved an impressive accuracy of 73% for overall survival (OS) prediction. [ABSTRACT FROM AUTHOR]

This narrative review explores the burgeoning field of Artificial Intelligence (AI)-driven Breast Cancer (BC) survival prediction, emphasizing the transformative impact on patient care. From machine learning to deep neural networks, diverse models demonstrate the potential to refine prognosis accuracy and tailor treatment strategies. The literature underscores the need for clinician integration and addresses challenges of model generalizability and ethical considerations. Crucially, AI's promise extends to Low- and Middle-Income Countries (LMICs), presenting an opportunity to bridge healthcare disparities. Collaborative efforts in research, technology transfer, and education are essential to empower healthcare professionals in LMICs. As we navigate this frontier, AI emerges not only as a technological advancement but as a guiding light toward personalized, accessible BC care, marking a significant stride in the global fight against this formidable disease. [ABSTRACT FROM AUTHOR]

Simple Summary: Colorectal cancer remains a major health challenge with high mortality rates and increasing diagnoses among younger adults. This study introduces a new method to predict patient survival using machine learning, a technique that can greatly enhance early diagnosis and treatment. The aim of our study is to apply eight different machine learning algorithms to a large data set of patients with colorectal cancer in Brazil and optimize these models with advanced parameter tuning tools. The best performing models achieved around 80% accuracy in predicting survival rates over one, three, and five years, as well as overall and cancer-specific mortality. This approach promises to improve clinical decision making by providing more accurate survival predictions, ultimately helping better patient care and management. Background: Colorectal cancer is one of the most prevalent forms of cancer and is associated with a high mortality rate. Additionally, an increasing number of adults under 50 are being diagnosed with the disease. This underscores the importance of leveraging modern technologies, such as artificial intelligence, for early diagnosis and treatment support. Methods: Eight classifiers were utilized in this research: Random Forest, XGBoost, CatBoost, LightGBM, Gradient Boosting, Extra Trees, the k-nearest neighbor algorithm (KNN), and decision trees. These algorithms were optimized using the frameworks Optuna, RayTune, and HyperOpt. This study was conducted on a public dataset from Brazil, containing information on tens of thousands of patients. Results: The models developed in this study demonstrated high classification accuracy in predicting one-, three-, and five-year survival, as well as overall mortality and cancer-specific mortality. The CatBoost, LightGBM, Gradient Boosting, and Random Forest classifiers delivered the best performance, achieving an accuracy of approximately 80% across all the evaluated tasks. Conclusions: This research enabled the development of effective classification models that can be applied in clinical practice. [ABSTRACT FROM AUTHOR]

Background and Aim: In this study, a deep learning algorithm was used to predict the survival rate of colon cancer (CC) patients, and compared its performance with traditional Cox regression. Methods: In this population‐based cohort study, we used the characteristics of patients diagnosed with CC between 2010 and 2015 from the Surveillance, Epidemiology and End Results (SEER) database. The population was randomized into a training set (n = 10 596, 70%) and a test set (n = 4536, 30%). Brier scores, area under the (AUC) receiver operating characteristic curve and calibration curves were used to compare the performance of the three most popular deep learning models, namely, artificial neural networks (ANN), deep neural networks (DNN), and long‐short term memory (LSTM) neural networks with Cox proportional hazard (CPH) model. Results: In the independent test set, the Brier values of ANN, DNN, LSTM and CPH were 0.155, 0.149, 0.148, and 0.170, respectively. The AUC values were 0.906 (95% confidence interval [CI] 0.897–0.916), 0.908 (95% CI 0.899–0.918), 0.910 (95% CI 0.901–0.919), and 0.793 (95% CI 0.769–0.816), respectively. Deep learning showed superior promising results than CPH in predicting CC specific survival. Conclusions: Deep learning showed potential advantages over traditional CPH models in terms of prognostic assessment and treatment recommendations. LSTM exhibited optimal predictive accuracy and has the ability to provide reliable information on individual survival and treatment recommendations for CC patients. [ABSTRACT FROM AUTHOR]

Purpose: This study was planned to determine the problems and affecting factors that children encounter Purpose: Colorectal cancer (CRC) is a leading cause of cancer-related mortality worldwide. Accurate survival prediction is crucial for advanced-stage patients to optimize treatment strategies and improve clinical outcomes. This study aimed to develop an artificial intelligence-assisted clinical decision support system (CDSS) for survival prediction in CRC patients using clinical and genomic data from the Cancer Genome Atlas Colon Adenocarcinoma Collection (TCGA-COAD) dataset. Methods: Machine learning algorithms, including C4.5 Decision Tree, Support Vector Machines (SVM), Random Forest, and Naive Bayes, were employed to create survival prediction models. Clinical parameters and genomic data from key pathways, such as glycolysis/gluconeogenesis and mTORCI, were integrated into the models. The models were evaluated based on accuracy and performance. Results: The Random Forest algorithm achieved the highest accuracy (82.3%) when only clinical parameters were used. When clinical data were combined with gene expression data, the model's accuracy increased further. The resulting models were incorporated into a user-friendly web interface, SurvCOCA, for clinical use. Conclusions: This study demonstrates the potential of Al-based tools to improve prognosis predictions in CRC patients. Further research is needed, with larger datasets and additional machine learning algorithms, to enhance clinical decision-making and optimize treatment strategies. [ABSTRACT FROM AUTHOR]

Cervical cancer is a highly prevalent malignancy affecting women worldwide, ranking as the seventh most common cancer globally. This study aims to systematically review and analyze cervical cancer survival predictions using machine learning (ML) algorithms. A comprehensive search was conducted across Scopus and PubMed databases in February 2024. Extracted articles were screened using Hubmeta software, with duplicates and non-relevant studies excluded. The final selection, comprising 24 articles, focused on survival predictions through ML techniques. These studies, published mostly post-2019, included datasets ranging from 75 to 9,462 cervical cancer patients and up to 91,294 squamous cell samples. The most commonly applied ML models were Random Forest (RF), Neural Networks (NN), Support Vector Machines (SVM), Ensemble and Hybrid Learning, and Deep Learning (DL). The area under the curve (AUC) for these models ranged from 0.84 to 0.9875, demonstrating their strong predictive capabilities. Clinical patient records were the primary data source. Meta-analysis was performed on the extracted data using GraphPad Prism for descriptive statistics and One-Way ANOVA. No significant differences were found between group means, as evidenced by an R-squared value of 0.1459. This result indicates that the independent variable (year of study) explained only 14.59% of the variance in ML model performance. The study found that the use of ML models has increased over time, particularly with Convolutional Neural Networks (CNNs) such as the ResNet50 model, which demonstrated superior accuracy metrics, including over 90% accuracy for the ResNet152 variant. These findings suggest that integrating multi-dimensional data with ML models holds significant potential for improving survival predictions in cervical cancer patients. Future research is recommended to develop tailored ML algorithms with even higher predictive accuracy for cervical cancer survival. [ABSTRACT FROM AUTHOR]

Heart failure (HF) is associated with high rates of morbidity and mortality. The value of deep learning survival prediction models using chest radiographs in patients with heart failure is currently unclear. The aim of our study is to develop and validate a deep learning survival prediction model using chest X-ray (DLSPCXR) in patients with HF. The study retrospectively enrolled a cohort of 353 patients with HF who underwent chest X-ray (CXR) at our institution between March 2012 and March 2017. The dataset was randomly divided into training (n = 247) and validation (n = 106) datasets. Univariate and multivariate Cox analysis were conducted on the training dataset to develop clinical and imaging survival prediction models. The DLSPCXR was trained and the selected clinical parameters were incorporated into DLSPCXR to establish a new model called DLSPinteg. Discrimination performance was evaluated using the time-dependent area under the receiver operating characteristic curves (TD AUC) at 1, 3, and 5-years survival. Delong's test was employed for the comparison of differences between two AUCs of different models. The risk-discrimination capability of the optimal model was evaluated by the Kaplan–Meier curve. In multivariable Cox analysis, older age, higher N-terminal pro-B-type natriuretic peptide (NT-ProBNP), systolic pulmonary artery pressure (sPAP) > 50 mmHg, New York Heart Association (NYHA) functional class III–IV and cardiothoracic ratio (CTR) ≥ 0.62 in CXR were independent predictors of poor prognosis in patients with HF. Based on the receiver operating characteristic (ROC) curve analysis, DLSPCXR had better performance at predicting 5-year survival than the imaging Cox model in the validation cohort (AUC: 0.757 vs. 0.561, P = 0.01). DLSPinteg as the optimal model outperforms the clinical Cox model (AUC: 0.826 vs. 0.633, P = 0.03), imaging Cox model (AUC: 0.826 vs. 0.555, P < 0.001), and DLSPCXR (AUC: 0.826 vs. 0.767, P = 0.06). Deep learning models using chest radiographs can predict survival in patients with heart failure with acceptable accuracy. [ABSTRACT FROM AUTHOR]

Abstract Background The rate of 30-day all-cause hospital readmissions can affect the funding a hospital receives. An accurate and reliable readmission prediction model could save money and increase quality-of-care. Few projects have explored formulating this task as a survival prediction problem, where models can exploit a real-valued time-to-readmission target. This paper demonstrates the effectiveness of a survival-inspired readmission model, especially when paired with a longitudinal patient representation that is agnostic to disease-cohort and predictive task. Methods We forecast readmissions for a population-level cohort of 421,088 patients discharged in 2015 and 2016 from hospitals in Alberta, Canada. Clinical features and sequences of historical medical codes (calculated from at least four full years prior to discharge) from linked administrative sources serve as model inputs. We trained binary 30-day readmission models (XGBoost and a Deep Neural Network) and time-to-event readmission models (CoxPH and N-MTLR) with and without machine-learned medical knowledge at initialization, then compared against the popular LACE-based model using the AUROC score at 30 days (AUROC@30). Survival models are additionally evaluated using concordance, Integrated Brier, and L1-loss scores. Results All models that utilize sequence features markedly out-perform even the best models trained on only clinical features. Further, a time-to-event target improves predictive performance at 30 days, given the same model inputs and architecture. N-MTLR, using solely sequence inputs and initialized with pre-learned medical knowledge, achieves an average AUROC@30 of 0.8460 over five folds with a standard deviation of 0.003. All trained models match or out-perform the LACE baseline of 0.6587±0.003. Conclusion Sequences of administrative medical codes contain rich predictive information for forecasting readmissions, and embedding medical knowledge a priori using machine learning provides readmission models an advantageous foundation for training. When combined with a model that can leverage a time-to-event target, excellent performance is possible on the 30-day all-cause readmission task using only administrative data.

Abstract Ovarian cancer is a formidable health challenge that demands accurate and timely survival predictions to guide clinical interventions. Existing methods, while commendable, suffer from limitations in harnessing the temporal evolution of patient data and capturing intricate interdependencies among different data elements. In this paper, we present a novel methodology which combines Temporal Analysis and Graph Neural Networks (GNNs) to significantly enhance ovarian cancer survival rate predictions. The shortcomings of current processes originate from their disability to correctly seize the complex interactions amongst diverse scientific information units in addition to the dynamic modifications that arise in a affected person`s nation over time. By combining temporal information evaluation and GNNs, our cautioned approach overcomes those drawbacks and, whilst as compared to preceding methods, yields a noteworthy 8.3% benefit in precision, 4.9% more accuracy, 5.5% more advantageous recall, and a considerable 2.9% reduction in prediction latency. Our method’s Temporal Analysis factor uses longitudinal affected person information to perceive good-sized styles and tendencies that offer precious insights into the direction of ovarian cancer. Through the combination of GNNs, we offer a robust framework able to shoot complicated interactions among exclusive capabilities of scientific data, permitting the version to realize diffused dependencies that would affect survival results. Our paintings have tremendous implications for scientific practice. Prompt and correct estimation of the survival price of ovarian most cancers allows scientific experts to customize remedy regimens, manipulate assets efficiently, and provide individualized care to patients. Additionally, the interpretability of our version`s predictions promotes a collaborative method for affected person care via way of means of strengthening agreement among scientific employees and the AI-driven selection help system. The proposed approach not only outperforms existing methods but also has the possible to develop ovarian cancer treatment by providing clinicians through a reliable tool for informed decision-making. Through a fusion of Temporal Analysis and Graph Neural Networks, we conduit the gap among data-driven insights and clinical practice, proposing a capable opportunity for refining patient outcomes in ovarian cancer management operations.

Abstract Background The 3-variable number-of-risk-factors (NRF) model is a prognostic tool for patients undergoing palliative radiotherapy (PRT). However, there is little research on the NRF model for patients with painful non-bone-metastasis tumours treated with PRT, and the efficacy of the NRF model in predicting survival is unclear to date. Therefore, we aimed to assess the prognostic accuracy of a 3-variable NRF model in patients undergoing PRT for bone and non- bone-metastasis tumours. Methods This was a secondary analysis of studies on PRT for bone-metastasis (BM) and PRT for miscellaneous painful tumours (MPTs), including non-BM tumours. Patients were grouped in the NRF model and survival was compared between groups. Discrimination was evaluated using a time-independent C-index and a time-dependent area under the receiver operating characteristic curve (AUROC). A calibration curve was used to assess the agreement between predicted and observed survival. Results We analysed 485 patients in the BM group and 302 patients in the MPT group. The median survival times in the BM group for groups I, II, and III were 35.1, 10.1, and 3.3 months, respectively (P

Abstract Background Predicting an individual’s risk of death from COVID-19 is essential for planning and optimising resources. However, since the real-world mortality rate is relatively low, particularly in places like Hong Kong, this makes building an accurate prediction model difficult due to the imbalanced nature of the dataset. This study introduces an innovative application of graph convolutional networks (GCNs) to predict COVID-19 patient survival using a highly imbalanced dataset. Unlike traditional models, GCNs leverage structural relationships within the data, enhancing predictive accuracy and robustness. By integrating demographic and laboratory data into a GCN framework, our approach addresses class imbalance and demonstrates significant improvements in prediction accuracy. Methods The cohort included all consecutive positive COVID-19 patients fulfilling study criteria admitted to 42 public hospitals in Hong Kong between January 23 and December 31, 2020 (n = 7,606). We proposed the population-based graph convolutional neural network (GCN) model which took blood test results, age and sex as inputs to predict the survival outcomes. Furthermore, we compared our proposed model to the Cox Proportional Hazard (CPH) model, conventional machine learning models, and oversampling machine learning models. Additionally, a subgroup analysis was performed on the test set in order to acquire a deeper understanding of the relationship between each patient node and its neighbours, revealing possible underlying causes of the inaccurate predictions. Results The GCN model was the top-performing model, with an AUC of 0.944, considerably outperforming all other models (p

Abstract Deep learning (DL) has shown potential to provide powerful representations of bulk RNA-seq data in cancer research. However, there is no consensus regarding the impact of design choices of DL approaches on the performance of the learned representation, including the model architecture, the training methodology and the various hyperparameters. To address this problem, we evaluate the performance of various design choices of DL representation learning methods using TCGA and DepMap pan-cancer datasets and assess their predictive power for survival and gene essentiality predictions. We demonstrate that baseline methods achieve comparable or superior performance compared to more complex models on survival predictions tasks. DL representation methods, however, are the most efficient to predict the gene essentiality of cell lines. We show that auto-encoders (AE) are consistently improved by techniques such as masking and multi-head training. Our results suggest that the impact of DL representations and of pretraining are highly task- and architecture-dependent, highlighting the need for adopting rigorous evaluation guidelines. These guidelines for robust evaluation are implemented in a pipeline made available to the research community.

Abstract The tumor microenvironment (TME) plays a fundamental role in tumorigenesis, tumor progression, and anti-cancer immunity potential of emerging cancer therapeutics. Understanding inter-patient TME heterogeneity, however, remains a challenge to efficient drug development. This article applies recent advances in machine learning (ML) for survival analysis to a retrospective study of NSCLC patients who received definitive surgical resection and immune pathology following surgery. ML methods are compared for their effectiveness in identifying prognostic subtypes. Six survival models, including Cox regression and five survival machine learning methods, were calibrated and applied to predict survival for NSCLC patients based on PD-L1 expression, CD3 expression, and ten baseline patient characteristics. Prognostic subregions of the biomarker space are delineated for each method using synthetic patient data augmentation and compared between models for overall survival concordance. A total of 423 NSCLC patients (46% female; median age [inter quantile range]: 67 [60–73]) treated with definite surgical resection were included in the study. And 219 (52%) patients experienced events during the observation period consisting of a maximum follow-up of 10 years and median follow up 78 months. The random survival forest (RSF) achieved the highest predictive accuracy, with a C-index of 0.84. The resultant biomarker subtypes demonstrate that patients with high PD-L1 expression combined with low CD3 counts experience higher risk of death within five-years of surgical resection.

Abstract Medulloblastoma is a malignant neuroepithelial tumor of the central nervous system. Accurate prediction of prognosis is essential for therapeutic decisions in medulloblastoma patients. We analyzed data from 2,322 medulloblastoma patients using the SEER database and randomly divided the dataset into training and testing datasets in a 7:3 ratio. We chose three models to build, one based on neural networks (DeepSurv), one based on ensemble learning that Random Survival Forest (RSF), and a typical Cox Proportional-hazards (CoxPH) model. The DeepSurv model outperformed the RSF and classic CoxPH models with C-indexes of 0.751 and 0.763 for the training and test datasets. Additionally, the DeepSurv model showed better accuracy in predicting 1-, 3-, and 5-year survival rates (AUC: 0.767–0.793). Therefore, our prediction model based on deep learning algorithms can more accurately predict the survival rate and survival period of medulloblastoma compared to other models.

Abstract Osteosarcoma is a primary malignant tumor that commonly affects children and adolescents, with a poor prognosis. The existence of tumor heterogeneity leads to different molecular subtypes and survival outcomes. Recently, lipid metabolism has been identified as a critical characteristic of cancer. Therefore, our study aims to identify osteosarcoma's lipid metabolism molecular subtype and develop a signature for survival outcome prediction. Four multicenter cohorts—TARGET-OS, GSE21257, GSE39058, and GSE16091—were amalgamated into a unified Meta-Cohort. Through consensus clustering, novel molecular subtypes within Meta-Cohort patients were delineated. Subsequent feature selection processes, encompassing analyses of differentially expressed genes between subtypes, univariate Cox analysis, and StepAIC, were employed to pinpoint biomarkers related to lipid metabolism in TARGET-OS. We selected the most effective algorithm for constructing a Lipid Metabolism-Related Signature (LMRS) by utilizing four machine-learning algorithms reconfigured into ten unique combinations. This selection was based on achieving the highest concordance index (C-index) in the test cohort of GSE21257, GSE39058, and GSE16091. We identified two distinct lipid metabolism molecular subtypes in osteosarcoma patients, C1 and C2, with significantly different survival rates. C1 is characterized by increased cholesterol, fatty acid synthesis, and ketone metabolism. In contrast, C2 focuses on steroid hormone biosynthesis, arachidonic acid, and glycerolipid and linoleic acid metabolism. Feature selection in the TARGET-OS identified 12 lipid metabolism genes, leading to a model predicting osteosarcoma patient survival. The LMRS, based on the 12 identified genes, consistently accurately predicted prognosis across TARGET-OS, testing cohorts, and Meta-Cohort. Incorporating 12 published signatures, LMRS showed robust and significantly superior predictive capability. Our results offer a promising tool to enhance the clinical management of osteosarcoma, potentially leading to improved clinical outcomes.

Background: Predicting an individual's risk of death from COVID-19 is essential for planning and optimising resources. However, since the real-world mortality rate is relatively low, particularly in places like Hong Kong, this makes building an accurate prediction model difficult due to the imbalanced nature of the dataset. This study introduces an innovative application of graph convolutional networks (GCNs) to predict COVID-19 patient survival using a highly imbalanced dataset. Unlike traditional models, GCNs leverage structural relationships within the data, enhancing predictive accuracy and robustness. By integrating demographic and laboratory data into a GCN framework, our approach addresses class imbalance and demonstrates significant improvements in prediction accuracy. Methods: The cohort included all consecutive positive COVID-19 patients fulfilling study criteria admitted to 42 public hospitals in Hong Kong between January 23 and December 31, 2020 (n = 7,606). We proposed the population-based graph convolutional neural network (GCN) model which took blood test results, age and sex as inputs to predict the survival outcomes. Furthermore, we compared our proposed model to the Cox Proportional Hazard (CPH) model, conventional machine learning models, and oversampling machine learning models. Additionally, a subgroup analysis was performed on the test set in order to acquire a deeper understanding of the relationship between each patient node and its neighbours, revealing possible underlying causes of the inaccurate predictions. Results: The GCN model was the top-performing model, with an AUC of 0.944, considerably outperforming all other models (p < 0.05), including the oversampled CPH model (0.708), linear regression (0.877), Linear Discriminant Analysis (0.860), K-nearest neighbours (0.834), Gaussian predictor (0.745) and support vector machine (0.847). With Kaplan-Meier estimates, the GCN model demonstrated good discriminability between low- and high-risk individuals (p < 0.0001). Based on subanalysis using the weighted-in score, although the GCN model was able to discriminate well between different predicted groups, the separation was inadequate between false negative (FN) and true negative (TN) groups. Conclusion: The GCN model considerably outperformed all other machine learning methods and baseline CPH models. Thus, when applied to this imbalanced COVID survival dataset, adopting a population graph representation may be an approach to achieving good prediction. [ABSTRACT FROM AUTHOR]

Deep learning (DL) has shown potential to provide powerful representations of bulk RNA-seq data in cancer research. However, there is no consensus regarding the impact of design choices of DL approaches on the performance of the learned representation, including the model architecture, the training methodology and the various hyperparameters. To address this problem, we evaluate the performance of various design choices of DL representation learning methods using TCGA and DepMap pan-cancer datasets and assess their predictive power for survival and gene essentiality predictions. We demonstrate that baseline methods achieve comparable or superior performance compared to more complex models on survival predictions tasks. DL representation methods, however, are the most efficient to predict the gene essentiality of cell lines. We show that auto-encoders (AE) are consistently improved by techniques such as masking and multi-head training. Our results suggest that the impact of DL representations and of pretraining are highly task- and architecture-dependent, highlighting the need for adopting rigorous evaluation guidelines. These guidelines for robust evaluation are implemented in a pipeline made available to the research community. [ABSTRACT FROM AUTHOR]

Simple Summary: Predicting the survival of bladder cancer patients following cystectomy can offer valuable information for treatment planning, decision-making, patient counseling, and resource allocation. Our aim was to develop large language model (LLM)-aided multi-modal predictive models, based on clinical information and CT images. These models achieved performances comparable to those of multi-modal predictive models that rely on manually extracted clinical information. This study demonstrates the potential of employing LLMs to process medical data, and of integrating LLM-processed data into modeling for prognosis. Survival prediction post-cystectomy is essential for the follow-up care of bladder cancer patients. This study aimed to evaluate artificial intelligence (AI)-large language models (LLMs) for extracting clinical information and improving image analysis, with an initial application involving predicting five-year survival rates of patients after radical cystectomy for bladder cancer. Data were retrospectively collected from medical records and CT urograms (CTUs) of bladder cancer patients between 2001 and 2020. Of 781 patients, 163 underwent chemotherapy, had pre- and post-chemotherapy CTUs, underwent radical cystectomy, and had an available post-surgery five-year survival follow-up. Five AI-LLMs (Dolly-v2, Vicuna-13b, Llama-2.0-13b, GPT-3.5, and GPT-4.0) were used to extract clinical descriptors from each patient's medical records. As a reference standard, clinical descriptors were also extracted manually. Radiomics and deep learning descriptors were extracted from CTU images. The developed multi-modal predictive model, CRD, was based on the clinical (C), radiomics (R), and deep learning (D) descriptors. The LLM retrieval accuracy was assessed. The performances of the survival predictive models were evaluated using AUC and Kaplan–Meier analysis. For the 163 patients (mean age 64 ± 9 years; M:F 131:32), the LLMs achieved extraction accuracies of 74%~87% (Dolly), 76%~83% (Vicuna), 82%~93% (Llama), 85%~91% (GPT-3.5), and 94%~97% (GPT-4.0). For a test dataset of 64 patients, the CRD model achieved AUCs of 0.89 ± 0.04 (manually extracted information), 0.87 ± 0.05 (Dolly), 0.83 ± 0.06~0.84 ± 0.05 (Vicuna), 0.81 ± 0.06~0.86 ± 0.05 (Llama), 0.85 ± 0.05~0.88 ± 0.05 (GPT-3.5), and 0.87 ± 0.05~0.88 ± 0.05 (GPT-4.0). This study demonstrates the use of LLM model-extracted clinical information, in conjunction with imaging analysis, to improve the prediction of clinical outcomes, with bladder cancer as an initial example. [ABSTRACT FROM AUTHOR]

This article discusses the value of serial assessments and the challenges of observational research in the context of extracorporeal cardiopulmonary resuscitation (ECPR) for patients with out-of-hospital cardiac arrest (OHCA). It highlights three large randomized controlled trials that evaluated the effect of ECPR on survival, each with different findings. The article then introduces a new study that examines the added value of serial lactate measurements for outcome prediction among adult patients undergoing ECPR after OHCA. The study finds that increased lactate clearance is associated with higher survival rates and more favorable neurological outcomes. However, the study has limitations, such as not incorporating arterial oxygen tension (Pao2) into the analysis, which may have influenced the results. Overall, the study emphasizes the importance of serial assessments and post-resuscitation care in improving outcomes for ECPR patients. [Extracted from the article]

Lung cancer, the second most prevalent form of cancer with the highest mortality rate, necessitates the stratification of patients based on their survival rates to develop effective treatment strategies. This study presents a two-stage framework for predicting lung cancer survival. The initial stage, classification, focuses on forecasting the five-year survival probability of lung cancer patients. Subsequent analysis was conducted on patients accurately classified as deceased during this stage. The second stage, regression, predicts the actual survival duration in months for deceased patients. This analysis employs the widely recognized Surveillance, Epidemiology, and End Results (SEER) database. To reduce dimensionality, two feature selection techniques, Recursive Feature Elimination with Random Forest (RFE-RF) and the Least Absolute Shrinkage and Selection Operator (LASSO), were adopted. Machine learning models were then trained using five-fold cross-validation for both classification and regression. Experimental results demonstrate that ensemble methods outperform other algorithms, including Logistic Regression (LR), Random Forest (RF), Multilayer Perceptron (MLP), Adaboost, and Naïve Bayes (NB), in terms of performance metrics. The existing techniques offer high accuracy for shorter survival periods, particularly for survival times of up to 6 months. Notably, the Light Gradient Boosting Machine (LGBM) classifier combined with RFE-RF achieves the highest classification accuracy of 89.6% and an area under the receiver operating characteristic (ROC) curve (AUC) score of 92.03 for survival durations up to 11 months. In regression analysis, the LGBM regressor outperforms its counterparts with a Mean Absolute Error (MAE) value of 7.53 and a Root Mean Squared Error (RMSE) value of 10.49. The study critically evaluates various cost functions' effectiveness in regression, validating the accuracy of survival duration predictions for the given dataset. [ABSTRACT FROM AUTHOR]

Current medical limitations in predicting cancer survival status and time necessitate advancements beyond traditional methods and physical indicators. This research introduces a novel two-stage prognostic framework for renal cell carcinoma, addressing the inadequacies of existing diagnostic approaches. In the first stage, the framework accurately predicts the survival status (alive or deceased) with metrics Accuracy, Precision, Recall, and F1 score to evaluate the effects of the classification results, while the second stage focuses on forecasting the future survival time of deceased patients with Root Mean Square Error and Mean Absolute Error to evaluate the regression results. Leveraging popular machine learning models, such as Adaptive Boosting, Extra Trees, Gradient Boosting, Random Forest, and Extreme Gradient Boosting, along with fusion models like Voting, Stacking, and Blending, our approach significantly improves prognostic accuracy as shown in our experiments. The novelty of our research lies in the integration of a logistic regression meta-model for interpreting the blending model's predictions, enhancing transparency. By the SHapley Additive exPlanations' interpretability, we provide insights into variable contributions, aiding understanding at both global and local levels. Through modal segmentation and multimodal fusion applied to raw data from the Surveillance, Epidemiology, and End Results program, we enhance the precision of renal cell carcinoma prognosis. Our proposed model provides an interpretable analysis of model predictions, highlighting key variables influencing classification and regression decisions in the two-stage renal cell carcinoma prognosis framework. By addressing the black-box problem inherent in machine learning, our proposed model helps healthcare practitioners with a more reliable and transparent basis for applying machine learning in cancer prognostication. [ABSTRACT FROM AUTHOR]

Medulloblastoma is a malignant neuroepithelial tumor of the central nervous system. Accurate prediction of prognosis is essential for therapeutic decisions in medulloblastoma patients. We analyzed data from 2,322 medulloblastoma patients using the SEER database and randomly divided the dataset into training and testing datasets in a 7:3 ratio. We chose three models to build, one based on neural networks (DeepSurv), one based on ensemble learning that Random Survival Forest (RSF), and a typical Cox Proportional-hazards (CoxPH) model. The DeepSurv model outperformed the RSF and classic CoxPH models with C-indexes of 0.751 and 0.763 for the training and test datasets. Additionally, the DeepSurv model showed better accuracy in predicting 1-, 3-, and 5-year survival rates (AUC: 0.767–0.793). Therefore, our prediction model based on deep learning algorithms can more accurately predict the survival rate and survival period of medulloblastoma compared to other models. [ABSTRACT FROM AUTHOR]

AbstractThis study attempted to build a prognostic riskscore model for pancreatic cancer (PC) patients based on vesicle-mediated transport protein-related genes (VMTGs). We initially conducted differential expression analysis and Cox regression analysis, followed by the construction of a riskscore model to classify PC patients into high-risk (HR) and low-risk (LR) groups. The GEO GSE62452 dataset further validated the model. Kaplan-Meier survival analysis was employed to analyze the survival rate of the HR group and LR group. Cox analysis confirmed the independent prognostic ability of the riskscore model. Additionally, we evaluated immune status in both HR and LR groups, utilizing data from the GDSC database to predict drug response among PC patients. We identified six PC-specific genes from 724 VMTGs. Survival analysis revealed that the survival rate of the HR group was lower than that of the LR group (P<0.05). Cox analysis confirmed that the prognostic riskscore model could independently predict the survival status of PC patients (P<0.001). Immunological analysis revealed that the ESTIMATE score, immune score, and stroma score of the HR group were considerably lower than those of the LR group, and the tumor purity score of the HR group was higher. The IC50 values of Gemcitabine, Irinotecan, Oxaliplatin, and Paclitaxel in the LR group were considerably lower than those in the HR group (P<0.001). In summary, the VMTG-based prognostic riskscore model could stratify PC risk and effectively predict the survival of PC patients. [ABSTRACT FROM AUTHOR]

Currently available risk scores fail to accurately predict morbidity and mortality in patients with severe symptomatic aortic stenosis who undergo transcatheter aortic valve implantation (TAVI). In this context, biomarkers like matrix metalloproteinase-2 (MMP-2) and Galectin-3 (Gal-3) may provide additional prognostic information. Patients with severe aortic stenosis undergoing consecutive, elective, transfemoral TAVI were included. Baseline demographic data, functional status, echocardiographic findings, clinical outcomes and biomarker levels were collected and analysed. The study cohort consisted of 89 patients (age 80.4 ± 5.1 years, EuroScore II 7.1 ± 5.8%). During a median follow-up period of 526 d, 28 patients (31.4%) died. Among those who died, median baseline MMP-2 (alive: 221.6 [170.4; 263] pg/mL vs. deceased: 272.1 [225; 308.8] pg/mL, p < 0.001) and Gal-3 levels (alive: 19.1 [13.5; 24.6] pg/mL vs. deceased: 25 [17.6; 29.5] pg/mL, p = 0.006) were higher than in survivors. In ROC analysis, MMP-2 reached an acceptable level of discrimination to predict mortality (AUC 0.733, 95% CI [0.62; 0.83], p < 0.001), but the predictive value of Gal-3 was poor (AUC 0.677, 95% CI [0.56; 0.79], p = 0.002). Kaplan–Meier and Cox regression analyses showed that patients with MMP-2 and Gal-3 concentrations above the median at baseline had significantly impaired long-term survival (p = 0.004 and p = 0.02, respectively). In patients with severe aortic stenosis undergoing transfemoral TAVI, MMP-2 and to a lesser extent Gal-3, seem to have additive value in optimizing risk prediction and streamlining decision-making. [ABSTRACT FROM AUTHOR]

Numerous factors have been associated with the survival outcomes in patients with spinal cord gliomas (SCG). Recognizing these specific determinants is crucial, yet it is also vital to establish a reliable and precise prognostic model for estimating individual survival outcomes. The objectives of this study are twofold: first, to create an array of interpretable machine learning (ML) models developed for predicting survival outcomes among SCG patients; and second, to integrate these models into an easily navigable online calculator to showcase their prospective clinical applicability. This was a retrospective, population-based cohort study aiming to predict the outcomes of interest, which were binary categorical variables, in SCG patients with ML models. The National Cancer Database (NCDB) was utilized to identify adults aged 18 years or older who were diagnosed with histologically confirmed SCGs between 2010 and 2019. The outcomes of interest were survival outcomes at three specific time points postdiagnosis: 1, 3, and 5 years. These outcomes were formed by combining the "Vital Status" and "Last Contact or Death (Months from Diagnosis)" variables. Model performance was evaluated visually and numerically. The visual evaluation utilized receiver operating characteristic (ROC) curves, precision-recall curves (PRCs), and calibration curves. The numerical evaluation involved metrics such as sensitivity, specificity, accuracy, area under the PRC (AUPRC), area under the ROC curve (AUROC), and Brier Score. We employed five ML algorithms—TabPFN, CatBoost, XGBoost, LightGBM, and Random Forest—along with the Optuna library for hyperparameter optimization. The models that yielded the highest AUROC values were chosen for integration into the online calculator. To enhance the explicability of our models, we utilized SHapley Additive exPlanations (SHAP) for assessing the relative significance of predictor variables and incorporated partial dependence plots (PDPs) to delineate the influence of singular variables on the predictions made by the top performing models. For the 1-year survival analysis, 4,913 patients [5.6% with 1-year mortality]; for the 3-year survival analysis, 4,027 patients (11.5% with 3-year mortality]; and for the 5-year survival analysis, 2,854 patients (20.4% with 5-year mortality) were included. The top models achieved AUROCs of 0.938 for 1-year mortality (TabPFN), 0.907 for 3-year mortality (LightGBM), and 0.902 for 5-year mortality (Random Forest). Global SHAP analyses across survival outcomes at different time points identified histology, tumor grade, age, surgery, radiotherapy, and tumor size as the most significant predictor variables for the top-performing models. This study demonstrates ML techniques can develop highly accurate prognostic models for SCG patients with excellent discriminatory ability. The interactive online calculator provides a tool for assessment by physicians (https://huggingface.co/spaces/MSHS-Neurosurgery-Research/NCDB-SCG). Local interpretability informs prediction influences for a given individual. External validation across diverse datasets could further substantiate potential utility and generalizability. This robust, interpretable methodology aligns with the goals of precision medicine, establishing a foundation for continued research leveraging ML's predictive power to enhance patient counseling. [ABSTRACT FROM AUTHOR]

The natural regeneration of forest ecosystems is crucial for their sustainability, but uncertainties have impeded the regeneration of some tree species. Identifying influencing factors and effective strategies to enhance seedling survival and growth is essential. We investigated factors affecting the natural regeneration of Larix principis‐rupprechtii and provided insights into seedling survival and growth. Eighteen artificial L. principis‐rupprechtii forest plots were established and monitored for 3 years. A logistic regression analysis and generalized linear models were used to investigate the influence of stand age, diameter at ground level, height, and other microhabitat factors on seedling regeneration. The microhabitat factors significantly influenced the overall L. principis‐rupprechtii regeneration density, as well as the density and growth of regenerated trees in different height classes. The area under the curve values for total nitrogen (0.796), total phosphorus (0.726), soil moisture (0.759), and litter thickness (0.633) were the highest, indicating a significant impact on the survival rate and mortality of the seedlings. Among these values, total nitrogen sensitivity (0.857) and specificity (0.810) were the highest, and the optimal threshold was 0.940. The survival rate decreased with increasing forest age, and the stands aged 4–7 years with a height of 1–2.5 m and a diameter at the ground level of approximately 2 cm constituted a relatively vulnerable and critical set of conditions for the survival of L. principis‐rupprechtii seedlings. The model showed that at 12 years old, L. principis‐rupprechtii trees were no longer vulnerable to mortality. The Kaplan–Meier model predicted future seedling survival through the construction of the comprehensive influence value and the measured seedling survival number. The model can be used to evaluate the survival rate for the final regeneration of a species, and targeted artificial seeding or replanting can improve the proportion of seedlings that survive. Our findings contribute to elucidating the factors affecting the natural regeneration of forest species and provide valuable insights for the development of effective regeneration strategies. [ABSTRACT FROM AUTHOR]