Search

Inconsistencies between performance data from laboratory‐prepared and field samples have been widely reported. These inconsistencies often result in inaccurate condition prediction, which leads to inefficient maintenance planning. Traditional pavement management systems (PMS) do not have the appropriate means (e.g., mechanistic solutions, extensive data handling facilities, etc.) to consider these data inconsistencies. With the growing demand for sustainable materials, there is a need for more self‐learning systems that could quickly transfer laboratory‐based information to field‐based information inside the PMS. The article aims to present a future‐ready machine learning‐based framework for analyzing the differences between laboratory and field‐prepared samples. Developed on the basis of data obtained from field and laboratory data, the gradient‐boosting decision trees‐based framework was able to establish a good relationship between laboratory performance and field performance (R2test > 80 for all models). At the same time, the framework could also show more complex relationships that are often not considered in practice. [ABSTRACT FROM AUTHOR]

Global declines in ecosystem extent and condition mean there is an increasing demand for recovery and conservation plans. Conservation plans for ecological communities require a management framework with measurable, time‐bound objectives. Efficient and structured processes that facilitate timely and comparable conservation plans are essential, especially where resources are constrained. We describe a process to streamline the development of conservation plans by combining functionally similar community sub‐types into a multi‐community State and Transition Model that can be used to guide conservation planning. We demonstrate this approach in a case study using eucalypt dominated woodlands of southern Australia—an ecosystem which occupies a vast geographical range across temperate Australia and includes many distinct vegetation communities, a growing number of which are endangered or threatened. Australian woodland ecologists (grouped according to their knowledge of three broad woodland sub‐types) were asked to develop causal chains to describe all factors associated with transitions among woodland condition states, and estimate the likelihoods associated with each transition at two time‐scales. The resultant State and Transition Model includes a set of eight general condition states that are common to eucalypt dominated woodlands, and 364 unique causal chains describing the drivers of all plausible transitions. We include an example of how the same information can be presented as a series of decision trees aimed at supporting on‐ground management decisions. The case study demonstrates that it is possible to construct a detailed State and Transition Model that synthesizes knowledge across multiple similar vegetation communities. To date, State and Transition Models focused on single communities or a smaller spatial scale, and this is the first attempt to construct a nationally relevant multi‐community State and Transition Model via a structured and participatory process. Synthesis and applications. This approach can be applied at multiple spatial scales to improve and streamline the development of robust conservation plans to improve how we plan for, implement and measure global biodiversity outcomes. [ABSTRACT FROM AUTHOR]

As the complexity of machine learning (ML) models increases and their application in different (and critical) domains grows, there is a strong demand for more interpretable and trustworthy ML. A direct, model‐agnostic, way to interpret such models is to train surrogate models—such as rule sets and decision trees—that sufficiently approximate the original ones while being simpler and easier‐to‐explain. Yet, rule sets can become very lengthy, with many if–else statements, and decision tree depth grows rapidly when accurately emulating complex ML models. In such cases, both approaches can fail to meet their core goal—providing users with model interpretability. To tackle this, we propose DeforestVis, a visual analytics tool that offers summarization of the behaviour of complex ML models by providing surrogate decision stumps (one‐level decision trees) generated with the Adaptive Boosting (AdaBoost) technique. DeforestVis helps users to explore the complexity versus fidelity trade‐off by incrementally generating more stumps, creating attribute‐based explanations with weighted stumps to justify decision making, and analysing the impact of rule overriding on training instance allocation between one or more stumps. An independent test set allows users to monitor the effectiveness of manual rule changes and form hypotheses based on case‐by‐case analyses. We show the applicability and usefulness of DeforestVis with two use cases and expert interviews with data analysts and model developers. [ABSTRACT FROM AUTHOR]

At present, data obtained from the Global Positioning System (GPS) is significantly valuable in mobility research. However, GPS‐based data lacks include trip purpose information. Consequently, many researchers have endeavoured to predict or impute these missing attributes. Existing studies have focused on constructing more features to improve prediction accuracy, but paid less attention to the model's applicability and transferability. In this study, five trip purposes are extracted, including education, recreation, personal, shopping, and transportation, from Chengdu Household Travel Survey (HTS) data. The individual and trip characteristics that are common and can be easily derived from GPS data are carefully selected and extracted. Point of Interest (POI) data of the trip destination are also collected to enhance input characteristics. To obtain more accurate results, an ensemble learning model, Gradient Boosting Decision Trees (GBDT), is employed to predict trip purposes. grid search and cross‐validation techniques are used to optimize the hyper‐parameters. Empirical results show that the proposed model achieves 0.788 accuracy, which is 22.17%, 14.53%, 10.36%, and 6.77% higher than Multinominal Logit (MNL), Artificial Neural Network (ANN), Random Forest (RF), and Deep Belief Network (DBN), respectively. It is also found that although increasing trip features improve the model's accuracy, it simultaneously impairs model's transferability and generalizability. [ABSTRACT FROM AUTHOR]

This study combines the use of multi‐objective decision trees (MODT) and a fuzzy TOPSIS approach to cope with sequential decision‐making under uncertainty with multiple conflicting objectives. To this aim, the MODT generates a Pareto‐optimal frontier of solutions for competing objectives, while fuzzy TOPSIS helps assess the noninferior solutions and assign weights to the objectives by considering possible uncertainties in outcomes as well as potential linguistic expression. The proposed approach is illustrated with a temporal decision problem concerning climate‐driven refugee resettlement, taking into consideration the challenges of global warming and system interdependencies in an uncertain context. The fuzzy TOPSIS approach was illustrated using two different procedures, and the procedure based on vector normalisation graphically depicts the fuzzy triangular representation of the final ranking. [ABSTRACT FROM AUTHOR]

Conducting high‐quality overviews of reviews (OoR) is time‐consuming. Because the quality of systematic reviews (SRs) varies, it is necessary to critically appraise SRs when conducting an OoR. A well‐established appraisal tool is A Measurement Tool to Assess Systematic Reviews (AMSTAR) 2, which takes about 15–32 min per application. To save time, we developed two fast‐and‐frugal decision trees (FFTs) for assessing the methodological quality of SR for OoR either during the full‐text screening stage (Screening FFT) or to the resulting pool of SRs (Rapid Appraisal FFT). To build a data set for developing the FFT, we identified published AMSTAR 2 appraisals. Overall confidence ratings of the AMSTAR 2 were used as a criterion and the 16 items as cues. One thousand five hundred and nineteen appraisals were obtained from 24 publications and divided into training and test data sets. The resulting Screening FFT consists of three items and correctly identifies all non‐critically low‐quality SRs (sensitivity of 100%), but has a positive predictive value of 59%. The three‐item Rapid Appraisal FFT correctly identifies 80% of the high‐quality SRs and correctly identifies 97% of the low‐quality SRs, resulting in an accuracy of 95%. The FFTs require about 10% of the 16 AMSTAR 2 items. The Screening FFT may be applied during full‐text screening to exclude SRs with critically low quality. The Rapid Appraisal FFT may be applied to the final SR pool to identify SR that might be of high methodological quality. [ABSTRACT FROM AUTHOR]

The exponential growth of digital text documents presents a significant challenge for text classification algorithms, as the vast number of words in each document can hinder their efficiency. Feature selection (FS) is a crucial technique that aims to eliminate irrelevant features and enhance classification accuracy. In this study, we propose an improved version of the discrete laying chicken algorithm (IDLCA) that utilizes noun‐based filtering to reduce the number of features and improve text classification performance. Although LCA is a newly proposed algorithm, it has not been systematically applied to discrete problems before. Our enhanced version of LCA employs different operators to improve both exploration and exploitation of this algorithm to find better solutions in discrete mode. To evaluate the effectiveness of the proposed method, we compared it with some conventional nature‐inspired feature selection methods using various learning models such as decision trees (DT), K‐nearest neighbor (KNN), Naive Bayes (NB), and support vector machine (SVM) on five benchmark datasets with three different evaluation metrics. The experimental results demonstrate the effectiveness of the proposed algorithm in comparison to the existing one. The code is available at https://github.com/m0javad/Improved-Discrete-Laying-Chicken-Algorithm. [ABSTRACT FROM AUTHOR]

Few studies included objective blood pressure (BP) to construct the predictive model of severe obstructive sleep apnea (OSA). This study used binary logistic regression model (BLRM) and the decision tree method (DTM) to constructed the predictive models for identifying severe OSA, and to compare the prediction capability between the two methods. Totally 499 adult patients with severe OSA and 1421 non‐severe OSA controls examined at the Sleep Medicine Center of a tertiary hospital in southern Taiwan between October 2016 and April 2019 were enrolled. OSA was diagnosed through polysomnography. Data on BP, demographic characteristics, anthropometric measurements, comorbidity histories, and sleep questionnaires were collected. BLRM and DTM were separately applied to identify predictors of severe OSA. The performance of risk scores was assessed by area under the receiver operating characteristic curves (AUCs). In BLRM, body mass index (BMI) ≥27 kg/m2, and Snore Outcomes Survey score ≤55 were significant predictors of severe OSA (AUC 0.623). In DTM, mean SpO2 <96%, average systolic BP ≥135 mmHg, and BMI ≥39 kg/m2 were observed to effectively differentiate cases of severe OSA (AUC 0.718). The AUC for the predictive models produced by the DTM was higher in older adults than in younger adults (0.807 vs. 0.723) mainly due to differences in clinical predictive features. In conclusion, DTM, using a different set of predictors, seems more effective in identifying severe OSA than BLRM. Differences in predictors ascertained demonstrated the necessity for separately constructing predictive models for younger and older adults. [ABSTRACT FROM AUTHOR]

Background: Chondrosarcoma (CHS), a bone malignancy, poses a significant challenge due to its heterogeneous nature and resistance to conventional treatments. There is a clear need for advanced prognostic instruments that can integrate multiple prognostic factors to deliver personalized survival predictions for individual patients. This study aimed to develop a novel prediction tool based on recursive partitioning analysis (RPA) to improve the estimation of overall survival for patients with CHS. Methods: Data from the Surveillance, Epidemiology, and End Results (SEER) database were analyzed, including demographic, clinical, and treatment details of patients diagnosed between 2000 and 2018. Using C5.0 algorithm, decision trees were created to predict survival probabilities at 12, 24, 60, and 120 months. The performance of the models was assessed through confusion scatter plot, accuracy rate, receiver operator characteristic (ROC) curve, and area under ROC curve (AUC). Results: The study identified tumor histology, surgery, age, visceral (brain/liver/lung) metastasis, chemotherapy, tumor grade, and sex as critical predictors. Decision trees revealed distinct patterns for survival prediction at each time point. The models showed high accuracy (82.40%–89.09% in training group, and 82.16%–88.74% in test group) and discriminatory power (AUC: 0.806–0.894 in training group, and 0.808–0.882 in test group) in both training and testing datasets. An interactive web‐based shiny APP (URL: https://yangxg1209.shinyapps.io/chondrosarcoma%5fsurvival%5fprediction/) was developed, simplifying the survival prediction process for clinicians. Conclusions: This study successfully employed RPA to develop a user‐friendly tool for personalized survival predictions in CHS. The decision tree models demonstrated robust predictive capabilities, with the interactive application facilitating clinical decision‐making. Future prospective studies are recommended to validate these findings and further refine the predictive model. [ABSTRACT FROM AUTHOR]

A variety of novel behavioral modeling techniques have been reported to accurately capture the nonlinear characteristics of GaN devices. For the purpose of describing GaN HEMT large‐signal behavior, machine learning‐based approaches have been proposed. There is, however, a lack of comparison and analysis of their performance with different machine learning techniques in these studies. An examination of machine learning‐based large‐signal modeling techniques is presented in this article. To develop large‐signal models of GaN HEMT, a range of commonly used modeling techniques such as artificial neural networks (ANN), random sample consistency (RANSAC), support vector regression (SVR), Gaussian process regression (GPR), decision trees (DTs), and genetic algorithm‐assisted ANN (GA‐ANN) are described and employed. Afterwards, integrated modeling techniques such as Bootstrap aggregation (BA), random forests (RF), AdaBoost, and gradient tree boosting (GTB) are reviewed and tested for their capabilities in developing GaN HEMT LSM. The hyperparameters of each built model are modified using the random search optimization (RSO) method, and five‐fold cross‐validation is used during validation. In order to identify optimal algorithms for nonlinear behavioral modeling of GaN devices, the model prediction results are comprehensively analyzed. [ABSTRACT FROM AUTHOR]

Background: As the population ages, mild cognitive impairment (MCI) and type 2 diabetes mellitus (T2DM) become common conditions that often coexist. Evidence has shown that MCI could lead to reduced treatment compliance, medication management, and self‐care ability in T2DM patients. Therefore, early identification of those with increased risk of MCI is crucial from a preventive perspective. Given the growing utilization of decision trees in prediction of health‐related outcomes, this study aimed to identify MCI in T2DM patients using the decision tree approach. Methods: This hospital‐based case–control study was performed in the Endocrinology Department of Xiangya Hospital affiliated to Central South University between March 2021 and December 2022. MCI was defined based on the Petersen criteria. Demographic characteristics, lifestyle factors, and T2DM‐related information were collected. The study sample was randomly divided into the training and validation sets in a 7:3 ratio. Univariate and multivariate analyses were performed, and a decision tree model was established using the chi‐square automatic interaction detection (CHAID) algorithm to identify key predictor variables associated with MCI. The area under the curve (AUC) value was used to evaluate the performance of the established decision tree model, and the performance of multivariate regression model was also evaluated for comparison. Results: A total of 1001 participants (705 in the training set and 296 in the validation set) were included in this study. The mean age of participants in the training and validation sets was 60.2 ± 10.3 and 60.4 ± 9.5 years, respectively. There were no significant differences in the characteristics between the training and validation sets (p >.05). The CHAID decision tree analysis identified six key predictor variables associated with MCI, including age, educational level, household income, regular physical activity, diabetic nephropathy, and diabetic retinopathy. The established decision tree model had 15 nodes composed of 4 layers, and age is the most significant predictor variable. It performed well (AUC =.75 [95% confidence interval (CI):.71–.78] and.67 [95% CI:.61–.74] in the training and validation sets, respectively), was internally validated, and had comparable predictive value compared to the multivariate logistic regression model (AUC =.76 [95% CI:.72–.80] and.69 [95% CI:.62–.75] in the training and validation sets, respectively). Conclusion: The established decision tree model based on age, educational level, household income, regular physical activity, diabetic nephropathy, and diabetic retinopathy performed well with comparable predictive value compared to the multivariate logistic regression model and was internally validated. Due to its superior classification accuracy and simple presentation as well as interpretation of collected data, the decision tree model is more recommended for the prediction of MCI in T2DM patients in clinical practice. [ABSTRACT FROM AUTHOR]

Soft robots have the advantage of adaptability and flexibility in various scenarios and tasks due to their inherent flexibility and mouldability, which makes them highly promising for real‐world applications. The development of electronic skin (E‐skin) perception systems is crucial for the advancement of soft robots. However, achieving both exteroceptive and proprioceptive capabilities in E‐skins, particularly in terms of decoupling and classifying sensing signals, remains a challenge. This study presents an E‐skin with mixed electronic and ionic conductivity that can simultaneously achieve exteroceptive and proprioceptive, based on the resistance response of conductive hydrogels. It is integrated with soft robots to enable state perception, with the sensed signals further decoded using the machine learning model of decision trees and random forest algorithms. The results demonstrate that the newly developed hydrogel sensing system can accurately predict attitude changes in soft robots when subjected to varying degrees of pressing, hot pressing, bending, twisting, and stretching. These findings that multifunctional hydrogels combine with machine learning to decode signals may serve as a basis for improving the sensing capabilities of intelligent soft robots in future advancements. [ABSTRACT FROM AUTHOR]

Background: Postoperative ileus (POI) continues to be a major cause of morbidity following colorectal surgery. Despite best efforts, the incidence of POI in colorectal surgery remains high (~30%). This study aimed to investigate machine learning techniques to identify risk factors for POI in colorectal surgery patients, to help guide further preventative strategies. Methods: A TRIPOD‐guideline‐compliant retrospective study was conducted for major colorectal surgery patients at a single tertial care centre (2018–2022). The primary outcome was the occurrence of POI, defined as not achieving GI‐2 (outcome measure of time to first stool and tolerance of oral diet) by day four. Multivariate logistic regression, decision trees, radial basis function and multilayer perceptron (MLP) models were trained using a random allocation of patients to training/testing data sets (80/20). The area under the receiver operating characteristic (AUROC) curves were used to evaluate model performance. Results: Of 504 colorectal surgery patients, 183 (36%) experienced POI. Multivariate logistic regression, decision trees, radial basis function and MLP models returned an AUROC of 0.722, 0.706, 0.712 and 0.800, respectively. The MLP model had the highest sensitivity and specificity values. In addition to well‐known risk factors for POI, such as postoperative hypokalaemia, surgical approach, and opioid use, the MLP model identified sarcopenia (ranked 4/30) as a potentially modifiable risk factor for POI. Conclusion: MLP outperformed other models in predicting POI. Machine learning can provide valuable insights into the importance and ranking of specific predictive variables for POI. Further research into the predictive value of preoperative sarcopenia for POI is required. [ABSTRACT FROM AUTHOR]

Objective: Artificial intelligence (AI)‐based decision support systems (DSS) are utilized in medicine but underlying decision‐making processes are usually unknown. Explainable AI (xAI) techniques provide insight into DSS, but little is known on how to design xAI for clinicians. Here we investigate the impact of various xAI techniques on a clinician's interaction with an AI‐based DSS in decision‐making tasks as compared to a general population. Methods: We conducted a randomized, blinded study in which members of the Child Neurology Society and American Academy of Neurology were compared to a general population. Participants received recommendations from a DSS via a random assignment of an xAI intervention (decision tree, crowd sourced agreement, case‐based reasoning, probability scores, counterfactual reasoning, feature importance, templated language, and no explanations). Primary outcomes included test performance and perceived explainability, trust, and social competence of the DSS. Secondary outcomes included compliance, understandability, and agreement per question. Results: We had 81 neurology participants with 284 in the general population. Decision trees were perceived as the more explainable by the medical versus general population (P < 0.01) and as more explainable than probability scores within the medical population (P < 0.001). Increasing neurology experience and perceived explainability degraded performance (P = 0.0214). Performance was not predicted by xAI method but by perceived explainability. Interpretation: xAI methods have different impacts on a medical versus general population; thus, xAI is not uniformly beneficial, and there is no one‐size‐fits‐all approach. Further user‐centered xAI research targeting clinicians and to develop personalized DSS for clinicians is needed. [ABSTRACT FROM AUTHOR]

Synthetic aperture radar (SAR) image change detection suffers from poor quality of the difference image and low detection accuracy. Hence, this paper proposes a SAR image change detection method based on a fused difference image and an optimized random forest scheme, termed LRN‐SSARF. Specifically, a fusion operator difference image LRN is proposed, which is generated using a weighted fusion of log‐ratio (LR), ratio (R), and normalized ratio (NoR). This difference image generation method reduces noise's influence. Then, the Otsu algorithm is applied to segment the difference image and select the training samples. The training samples are input into the random forest (RF) model optimised by the sparrow search algorithm (SSA) for training and classification. Finally, the region link is uesd to refine the detection results and generate the final result. The change detection results of six real SAR image scenes highlight that the proposed algorithm has a high detection accuracy, and affords appealing integrity and detailed information about the change regions. Specially, the detection accuracy advantage of the Bangladesh dataset is larger, with the accuracy and Kappa coefficient reaching 98.04% and 92.00%, much higher than the competitor methods. [ABSTRACT FROM AUTHOR]

An Objective Structured Practical Examination (OSPE) is an effective and robust, but resource‐intensive, means of evaluating anatomical knowledge. Since most OSPEs employ short answer or fill‐in‐the‐blank style questions, the format requires many people familiar with the content to mark the examinations. However, the increasing prevalence of online delivery for anatomy and physiology courses could result in students losing the OSPE practice that they would receive in face‐to‐face learning sessions. The purpose of this study was to test the accuracy of Decision Trees (DTs) in marking OSPE questions as a first step to creating an intelligent, online OSPE tutoring system. The study used the results of the winter 2020 semester final OSPE from McMaster University's anatomy and physiology course in the Faculty of Health Sciences (HTHSCI 2FF3/2LL3/1D06) as the data set. Ninety percent of the data set was used in a 10‐fold validation algorithm to train a DT for each of the 54 questions. Each DT was comprised of unique words that appeared in correct, student‐written answers. The remaining 10% of the data set was marked by the generated DTs. When the answers marked by the DT were compared to the answers marked by staff and faculty, the DT achieved an average accuracy of 94.49% across all 54 questions. This suggests that machine learning algorithms such as DTs are a highly effective option for OSPE grading and are suitable for the development of an intelligent, online OSPE tutoring system. [ABSTRACT FROM AUTHOR]

Despite today's developing healthcare technology, conventional colonoscopy is still a gold‐standard method to detect colon abnormalities. Due to the folded structure of the intestine and visual disturbances caused by artifacts, it can be hard for specialists to detect abnormalities during the procedure. Frames that include artifacts such as specular reflection, improper contrast levels from insufficient or excessive illumination gastric juice, bubbles, or residuals should be detected to increase an accurate diagnosis rate. In this work, both conventional machine learning and transfer learning methods have been used to detect non‐informative frames in colonoscopy videos. The conventional machine learning part consists of 5 different types of texture features, which are gray level co‐occurrence matrix (GLCM), gray level run length matrix (GLRLM), neighborhood gray‐tone difference matrix (NGTDM), focus measure operators (FMOs), and first‐order statistics. In addition to these methods, we utilized 8 different transfer learning models: AlexNet, SqueezeNet, GoogleNet, ShuffleNet, ResNet50, ResNet18, NasNetMobile, and MobileNet. The results showed that FMOs and decision tree combination gave the best accuracy and f‐measure values with almost 89% and 0.79%, respectively, for the conventional machine learning part. When the transfer learning part is taken into account, AlexNet (99.85%) and SqueezeNet (98.80%) have the highest performance metric results. This study shows the potential of both transfer learning and conventional machine learning algorithms to provide fast and accurate non‐informative frame detection to be used during a colonoscopy, which may be considered the initial step in identifying and classifying colon‐related diseases automatically to help guide physicians. [ABSTRACT FROM AUTHOR]

Objectives: To describe a new classification for intraoral scans based on the scan extension and to introduce a decision guideline to choose the scan extension for fabricating tooth‐ and implant‐supported fixed dental prostheses (FDPs). Overview: Multiple operator‐ and patient‐related factors have been identified that can decrease the scanning accuracy of intraoral scanners (IOSs), including scan extension. However, the decision criteria for selecting scan extension for fabricating tooth‐ and implant‐supported restorations is unclear. Based on the extension of the intraoral digital scans, three types of scans can be defined: half‐arch (anterior or posterior), extended half‐arch, and complete‐arch scan. Variables to consider when choosing the scan extension include the number and location of units being restored, as well as the extension and location of edentulous areas. Additionally, the accuracy of the virtual definitive cast and the accuracy of the maxillomandibular relationship captured by using IOSs should be differentiated. Conclusions: A decision tree for selecting the scan extension is presented. The decision is based on the number and location of units being restored, and the extension and location of edentulous areas. Intraoral scans with reduced scan extension are indicated when fabricating tooth‐ and implant‐supported crowns or short‐span fixed prostheses, when the patient does not have more than one missing tooth in the area of the dental arch included in the scan. For the remaining clinical conditions, complete‐arch intraoral scans are recommended. Clinical Significance: Scan extension is a clinician's decision that should be based on the number and location of units being restored and the extension and location of edentulous areas. Intraoral scans with a reduced scan extension is recommended, when possible. [ABSTRACT FROM AUTHOR]

In seismic data acquisition, because of several factors, such as surface barriers, receiver failure, noise contamination and budget control, seismic records often exhibit irregular sampling in the space domain. As corrupted seismic records have a negative effect on seismic migration, inversion and interpretation, seismic trace interpolation is a key step in seismic data pre‐processing. In this paper, we propose a high‐efficiency and high‐precision seismic trace interpolation method for irregularly spatially sampled data by combining an extreme gradient boosting decision tree and principal component analysis in a semi‐supervised learning method. The adjacent trace number, sampling number and amplitudes of the effective seismic data were taken as features to build the training data set for the extreme gradient boosting decision tree. Principal component analysis is applied to remove redundant information and accelerate the training speed. This is different from the traditional trace interpolation method in that the proposed method is data‐driven; therefore, it does not require any assumptions. Compared with other deep learning‐based trace interpolation methods, the proposed method has fewer control parameters and learning labels and a smaller training cost. Experiments using synthetic and field data demonstrated the validity and flexibility of this trace interpolation method. [ABSTRACT FROM AUTHOR]

Objective: We have developed explainable machine learning models to predict the overall survival (OS) of retroperitoneal liposarcoma (RLPS) patients. This approach aims to enhance the explainability and transparency of our modeling results. Methods: We collected clinicopathological information of RLPS patients from The Surveillance, Epidemiology, and End Results (SEER) database and allocated them into training and validation sets with a 7:3 ratio. Simultaneously, we obtained an external validation cohort from The First Affiliated Hospital of Naval Medical University (Shanghai, China). We performed LASSO regression and multivariate Cox proportional hazards analysis to identify relevant risk factors, which were then combined to develop six machine learning (ML) models: Cox proportional hazards model (Coxph), random survival forest (RSF), ranger, gradient boosting with component‐wise linear models (GBM), decision trees, and boosting trees. The predictive performance of these ML models was evaluated using the concordance index (C‐index), the integrated cumulative/dynamic area under the curve (AUC), and the integrated Brier score, as well as the Cox–Snell residual plot. We also used time‐dependent variable importance, analysis of partial dependence survival plots, and the generation of aggregated survival SHapley Additive exPlanations (SurvSHAP) plots to provide a global explanation of the optimal model. Additionally, SurvSHAP (t) and survival local interpretable model‐agnostic explanations (SurvLIME) plots were used to provide a local explanation of the optimal model. Results: The final ML models are consisted of six factors: patient's age, gender, marital status, surgical history, as well as tumor's histopathological classification, histological grade, and SEER stage. Our prognostic model exhibits significant discriminative ability, particularly with the ranger model performing optimally. In the training set, validation set, and external validation set, the AUC for 1, 3, and 5 year OS are all above 0.83, and the integrated Brier scores are consistently below 0.15. The explainability analysis of the ranger model also indicates that histological grade, histopathological classification, and age are the most influential factors in predicting OS. Conclusions: The ranger ML prognostic model exhibits optimal performance and can be utilized to predict the OS of RLPS patients, offering valuable and crucial references for clinical physicians to make informed decisions in advance. [ABSTRACT FROM AUTHOR]

Aims: To explore the clinico‐sero‐pathological characteristics and risk prediction model of idiopathic inflammatory myopathy (IIM) patients with different muscular perifascicular (PF) changes. Methods: IIM patients in our center were enrolled and the clinico‐sero‐pathological data were retrospectively analyzed. A decision tree model was established through machine learning. Results: There were 231 IIM patients enrolled, including 53 with perifascicular atrophy (PFA), 39 with perifascicular necrosis (PFN), and 26 with isolated perifascicular enhancement of MHC‐I/MHC‐II (PF‐MHCn). Clinically, PFA patients exhibited skin rashes and dermatomyositis‐specific antibodies (DM‐MSAs, 74.5%) except for anti‐Mi2. PFN patients showed the most severe muscle weakness, highest creatine kinase (CK), anti‐Mi2 (56.8%), and anti‐Jo‐1 (24.3%) antibodies. PF‐MHCn patients demonstrated negative MSAs (48.0%) and elevated CK. Histopathologically, MAC predominantly deposited on PF capillaries in PFA but on non‐necrotic myofiber in PFN (43.4% and 36.8%, p < 0.001). MxA expression was least in PF‐MHCn (36.0% vs. 83.0% vs. 63.2%, p < 0.001). The decision tree model could effectively predict different subgroups, especially PFA and PFN. Conclusions: Three types of PF change of IIMs representing distinct clinico‐serological characteristics and pathomechanism. Undiscovered MSAs should be explored especially in PF‐MHCn patients. The three pathological features could be accurately predicted through the decision tree model. [ABSTRACT FROM AUTHOR]

Cable trees are primarily employed in industrial products to facilitate energy transfer and information exchange among various components. When utilizing machines for assembly, it is essential to convert the wiring plan into a sequence of cable insertion operations executed by the machine under various constraints. This poses a combinatorial optimization problem. In this domain, constraint programming (CP) solvers often exhibit outstanding performance by leveraging their robust problem‐modelling capabilities, excellent scalability, and precise solving capabilities. However, CP solvers may achieve various performances for different problem instances. Selecting the most suitable CP solver for each problem instance is crucial. This paper introduces an automatic selection algorithm for CP solvers to solve the cable tree wiring problem (CTW). Firstly, a scoring system is used to conduct an in‐depth analysis and compare four well‐known CP solvers: CPLEX, Chuffed, OR‐Tools, and Gurobi. The results indicate that OR‐Tools and CPLEX outperform other solvers in performance. Moreover, these two solvers exhibit complementary advantages in quickly finding optimal and feasible solutions within specified time limits. Therefore, CP and machine learning are ingeniously integrated, harnessing their complementary advantages. 4240 instances covering various scenarios are randomly generated to form the problem space. This method incorporates decision trees, random forests, K‐nearest neighbours, and naive Bayes, utilizing these four machine learning techniques. The proposed method can achieve better results than traditional single CP solvers. Among all the evaluated machining learning techniques, the automatic solver selection methods based on decision trees and random forests can achieve accuracy rates of 91.29% and 84.15%, respectively. [ABSTRACT FROM AUTHOR]

BACKGROUND: Previously, a lexicon and protocol for quantitative descriptive analysis (QDA) was established for the Uganda sweetpotato breeding program. The implication of QDA scores for priority sensory attributes on consumer preference should be determined to interpret results efficiently and make decisions effectively. The present study aimed to develop a gender‐responsive decision tree to obtain an overall sweetpotato eating quality score to facilitate demand‐led targeted breeding selection. It focused on Kamuli and Hoima districts (Uganda) and uses pre‐lease advanced clones ('NKB3', 'NKB105', 'NKB135', 'D11' and 'D20'), released varieties ('NASPOT 8' and 'NAROSPOT 1') and landraces ('Muwulu‐Aduduma', 'Umbrella'). RESULTS: Including boiled sweetpotato sensory characteristics, namely mealy, sweet taste, sweetpotato smell, firm and not fibrous, in breeding design would benefit end‐users, especially women given their role in varietal selection, food preparation and marketing. 'D20', 'NASPOT 8' and 'NAROSPOT 1' were most liked in both districts. 'NKB3' and 'D11' were the least liked in Hoima, whereas 'Muwulu‐Aduduma' was the least liked in Kamuli. There was a positive correlation between color and overall liking (r2 = 0.8) and consumers liked the color (average rating ≥ 6 on a nine‐point hedonic scale) of all genotypes. Threshold values (average rating on 11‐point scales) for consumer acceptability were identified (sweet taste = 6, sweetpotato aroma and flavor = 6, firmness = 3, and mealiness = 4). A regression decision tree tool was created to calculate an eating quality selection index when screening lines in breeding programs using the values. CONCLUSION: Decision trees that include consumer needs and gender considerations would facilitate demand‐led breeding and make varietal selection in sweetpotato breeding programs more effective. © 2023 The Authors. Journal of The Science of Food and Agriculture published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry. [ABSTRACT FROM AUTHOR]

Background: Current methods for developing clinical practice guidelines have several limitations: they are characterised by the "black box" operation—a process with defined inputs and outputs but an incomplete understanding of its internal workings; they have "the integration problem"—a lack of framework for explicitly integrating factors such as patient preferences and trade‐offs between benefits and harms; they generate one recommendation at a time that typically are not connected in a coherent analytical framework; and they apply to "average" patients, while clinicians and their patients seek advice tailored to individual circumstances. Methods: We propose augmenting the current guideline development method by converting evidence‐based pathways into fast‐and‐frugal decision trees (FFTs) and integrating them with generalised decision curve analysis to formulate clear, individualised management recommendations. Results: We illustrate the process by developing recommendations for the management of heparin‐induced thrombocytopenia (HIT). We converted evidence‐based pathways for HIT, developed by the American Society of Hematology, into an FFT. Here, we consider only thrombotic complications and major bleeding. We leveraged the predictive potential of FFTs to compare the effects of argatroban, bivalirudin, fondaparinux, and direct oral anticoagulants (DOACs) using generalised decision curve analysis. We found that DOACs were superior to other treatments if the FFT‐predicted probability of HIT exceeded 3%. Conclusions: The proposed analytical framework connects guidelines, pathways, FFTs, and decision analysis, offering risk‐tailored personalised recommendations and addressing current guideline development critiques. [ABSTRACT FROM AUTHOR]

Purpose: Develop an efficient, interactive, and instructive checklist document for the management of implanted electronic medical devices in a multimodality radiotherapy clinic. Methods: The built-in scripting and interactivity of a popular commercial word processor was used to develop an interactive document that changes the information presented to the user based on drop-down selections. The interactivity and scripting were compatible with the radiation oncology information system (ROIS) which allows the document to be accessible by all team members and serve as a permanent record in a patient's electronic chart. Results: The final interactive document,which was clinically deployed after beta testing with a group consisting of nurses and medical physicists, presents information and action plans to the user based on multiple departmental medical device decision trees that are specific to the combination of device, treatment modality, rhythm-pacing dependence for cardiac devices, and distance from the device to the treatment volume. Conclusion: A script-enabled interactive document was developed for a busy multimodality clinic, condensing multiple comprehensive departmental guidelines spanning multiple device types and treatment modalities into a single interactive checklist accessible within the ROIS. Given the wide accessibility of the commercial word processor, this approach could be adopted by other clinics to streamline their own respective workflows. [ABSTRACT FROM AUTHOR]

The current study aims to determine the programming performance (low and high) of the students between the ages of 12–24 who receive programming education, by descriptive analysis, to determine the current situation according to various variables and to redict them with machine learning algorithms. Thus, the change in programming performance, computational identity, computational thinking perspective, programming empowerment, and programming anxiety according to various variables was examined. The performances of different algorithms were compared in estimating the low and high programming performance of these variables. The research involved 763 students who were between the ages of 12 and 24 and had received programming education. Different scales were used to collect the opinions of students who received programming education. Descriptive analyses, one‐way analysis of variance (ANOVA), and machine learning algorithms were used in the analysis of the data set. Analyzes were made in Statistical Package for the Social Sciences (SPSS) and Knime software. Decision trees, k‐nearest neighbors, support vector machines, random forest, Naive Bayes classifiers, and logistic regression were used in the study. Students' programming performance, computational identity, computational thinking perspective, and programming empowerment mean scores differ in terms of gender and educational level variables. Research variables do not differ statistically in terms of academic success level variables. According to the programming performance level, students' computational identity, computational thinking perspective, and programming empowerment scores differ. No significant difference was found between programming anxiety scores. Decision trees of the algorithm with the highest accuracy result according to low and high programming performance conditions are [(0.966), (0.966)], respectively. The fact that these obtained scores are above 90% can be interpreted as sufficient estimation performance. [ABSTRACT FROM AUTHOR]

Background: Nurses in neurointensive care units (NCUs) commonly use physical restraint (PR) to prevent adverse events like unplanned removal of devices (URDs) or falls. However, PR use should be based on evidenced decisions as it has drawbacks. Unfortunately, there is a lack of research‐based PR protocol to support decision‐making for nurses, especially for neurocritical patients. Aim: This study developed a restraint decision tree for neurocritical patients (RDT‐N) to assist nurses in making PR decisions. We assessed its effectiveness in reducing PR use and adverse events. Study Design: This study employed a baseline and post‐intervention test design at a NCU with 19 beds and 45 nurses in a tertiary hospital in a metropolitan city in South Korea. Two‐hundred and thirty‐seven adult patients were admitted during the study period. During the intervention, nurses were trained on the RDT‐N. PR use and adverse events between the baseline and post‐intervention periods were compared. Results: Post‐intervention, total number of restrained patients decreased (20.7%–16.3%; χ2 = 7.68, p =.006), and the average number of PR applied per restrained patient decreased (2.42–1.71; t = 5.74, p <.001). The most frequently used PR type changed from extremity cuff to mitten (χ2 = 397.62, p <.001). No falls occurred during the study periods. On the other hand, URDs at baseline were 18.67 cases per 1000 patient days in the high‐risk group and 5.78 cases per 1000 patient days in the moderate‐risk group; however, no URD cases were reported post‐intervention. Conclusions: The RDT‐N effectively reduced PR use and adverse events. Its application can enhance patient‐centred care based on individual condition and potential risks in NCUs. Relevance to Clinical Practice: Nurses can use the RDT‐N to assess the need for PR in caring for neurocritical patients, reducing PR use and adverse events. [ABSTRACT FROM AUTHOR]

Administration of a new drug candidate in a first‐in‐human (FIH) clinical trial is a particularly challenging phase in drug development and is especially true for immunomodulators, which are a diverse and complex class of drugs with a broad range of mechanisms of action and associated safety risks. Risk is generally greater for immunostimulators, in which safety concerns are associated with acute toxicity, compared to immunosuppressors, where the risks are related to chronic effects. Current methodologies for FIH dose selection for immunostimulators are focused primarily on identifying the minimum anticipated biological effect level (MABEL), which has often resulted in sub‐therapeutic doses, leading to long and costly escalation phases. The Health and Environmental Sciences Institute (HESI) – Immuno‐Safety Technical Committee (ITC) organized a project to address this issue through two complementary approaches: (i) an industry survey on FIH dose selection strategies and (ii) detailed case studies for immunomodulators in oncology and non‐oncology indications. Key messages from the industry survey responses highlighted a preference toward more dynamic PK/PD approaches as in vitro assays are seemingly not representative of true physiological conditions for immunomodulators. These principles are highlighted in case studies. To address the above themes, we have proposed a revised decision tree, which expands on the guidance by the IQ MABEL Working Group (Leach et al. 2021). This approach facilitates a more refined recommendation of FIH dose selection for immunomodulators, allowing for a nuanced consideration of their mechanisms of action (MOAs) and the associated risk‐to‐benefit ratio, among other factors. [ABSTRACT FROM AUTHOR]

The learning effect of students is crucial for assessing teaching quality, thus playing a significant role in teaching management. Predicting student achievement is a major challenge in understanding the learning effect of students. Currently, many studies have utilized machine learning methods such as the decision tree algorithms C4.5, ID3, CART, J48, random forest, and others. However, few studies have explored the use of the Bagging algorithm in this field. Therefore, this study proposes a classification prediction method for student achievement based on the Bagging‐CART algorithm. Initially, the student achievement data is preprocessed, and the Apriori method is applied to mine the strongly associated dataset. The optimal hyper‐parameters are determined through grid search to train and predict the Bagging‐CART algorithm. Furthermore, the CART, J48, and Bagging‐CART algorithms are trained, and their evaluation indicators are compared using a confusion matrix. The results indicate that the Bagging‐CART model achieves an accuracy of 98.16%, a recall rate of 91.80%, a precision of 90.83%, and an F1 score of 94.87%. In comparison, the accuracy, precision, and F1 scores are higher than those obtained with CART and J48. Although the recall rate is slightly lower than that of CART by 0.26%, it is 0.52% higher than that of J48. Consequently, this method demonstrates strong predictive capabilities and introduces a new reference method for evaluating students' learning effect. [ABSTRACT FROM AUTHOR]

Background: There is little evidence regarding risk prediction for surgical site infection (SSI) after lower third molar (L3M) surgery. Methods: We conducted a nested case–control study to develop a multivariable logistic model for predicting the risk of SSI after L3M surgery. Data were obtained from Hokkaido University Hospital from April 2013 to March 2020. Multiple imputation was applied for the missing values. We conducted decision tree (DT) analysis to evaluate the combinations of factors affecting SSI risk. Results: We identified 648 patients. The final model retained the available distal space (Pell & Gregory II [p = 0.05], Pell & Gregory III [p < 0.01]), depth (Pell & Gregory B [p < 0.01], Pell & Gregory C [p < 0.01]), surgeon's experience (3–10 years [p = 0.25], <3 years [p < 0.01]), and simultaneous extraction of both L3M [p < 0.01]; the concordance‐statistic was 0.72. The DT analysis demonstrated that patients with Pell and Gregory B or C and simultaneous extraction of both L3M had the highest risk of SSI. Conclusions: We developed a model for predicting SSI after L3M surgery with adequate predictive metrics in a single center. This model will make the SSI risk prediction more accessible. [ABSTRACT FROM AUTHOR]

Coordinated policies and effective global environmental governance are needed to address the global biodiversity crisis. Human dimensions like geopolitics influence conservation decision‐making and outcomes. The importance of considering these complex social factors is heightened in an era of renewed great‐power politics, as the intensifying US–China rivalry has direct implications for environmental governance and biodiversity outcomes. Can the US–China rivalry and its confrontational dynamics be channeled into conservation policymaking to improve biodiversity outcomes? Drawing from international relations and policy studies, policy diffusion theory can provide conservationists with useful insights into the interdependency of policy decisions. Here, we examine the four mechanisms—competition, coercion, learning, and emulation—of the classic model of policy diffusion theory in the context of environmental policymaking. We explore a case study for each mechanism to illustrate how it can benefit biodiversity conservation, and point to examples of relevant policies and actions that could improve outcomes. To operationalize this concept for conservation, we present a decision tree that conservationists can use to determine the most relevant policy diffusion mechanism in different policy contexts. Upon determining the appropriate mechanism, conservationists can take further steps to intentionally trigger the mechanism and catalyze conservation policy diffusion across jurisdictions. [ABSTRACT FROM AUTHOR]

Injury surveillance data indicate that collegiate athletes are at greater risk for lower extremity (LE) injuries following sports‐related concussion (SRC). While the association between SRC and LE injury appears to be clinically relevant up to 1‐year post‐SRC, little evidence has been provided to determine possible mechanistic rationales. Thus, we aimed to compare collegiate athletes with a history of SRC to matched controls on biomechanical and cognitive performance measures associated with LE injury risk. Athletes with a history of SRC (n = 20) and matched controls (n = 20) performed unanticipated bilateral land‐and‐cut tasks and cognitive assessments. Group‐based analyses (ANOVA) and predictive modeling (C5.0 decision tree algorithm) were used to compare group differences on biomechanical and cognitive measures. Collegiate athletes with a history of SRC demonstrated approximately six degrees less peak knee flexion on both dominant (p = 0.03, d = 0.71) and nondominant (p = 0.02, d = 0.78) limbs during the land‐and‐cut tasks compared to controls. Verbal Memory, knee flexion, and Go/No Go total score (C5.0 decision tree algorithm) were identified as the strongest indicators of previous SRC injury history. Reduced knee flexion during sport‐specific land‐and‐cut tasks may be a mechanism for increased LE injury risk in athletes with a history of SRC. There appears to be multiple biomechanical and cognitive predictors for identifying previous SRC in collegiate athletes, providing evidence to support a multifactorial SRC management strategy to reduce future injury risk. [ABSTRACT FROM AUTHOR]

Introduction: Death due to covid‐19 is one of the biggest health challenges in the world. There are many models that can predict death due to COVID‐19. This study aimed to fit and compare Decision Tree (DT), Support Vector Machine (SVM), and AdaBoost models to predict death due to COVID‐19. Methods: To describe the variables, mean (SD) and frequency (%) were reported. To determine the relationship between the variables and the death caused by COVID‐19, chi‐square test was performed with a significance level of 0.05. To compare DT, SVM and AdaBoost models for predicting death due to COVID‐19 from sensitivity, specificity, accuracy and the area under the rock curve under R software using psych, caTools, random over‐sampling examples, rpart, rpartplot packages was done. Results: Out of the total of 23,054 patients studied, 10,935 cases (46.5%) were women, and 12,569 cases (53.5%) were men. Additionally, the mean age of the patients was 54.9 ± 21.0 years. There is a statistically significant relationship between gender, fever, cough, muscle pain, smell and taste, abdominal pain, nausea and vomiting, diarrhea, anorexia, dizziness, chest pain, intubation, cancer, diabetes, chronic blood disease, Violation of immunity, pregnancy, Dialysis, chronic lung disease with the death of covid‐19 patients showed (p < 0.05). The results showed that the sensitivity, specificity, accuracy and the area under the receiver operating characteristic curve were respectively 0.60, 0.68, 0.71, and 0.75 in the DT model, 0.54, 0.62, 0.63, and 0.71 in the SVM model, and 0.59, 0.65, 0.69 and 0.74 in the AdaBoost model. Conclusion: The results showed that DT had a high predictive power compared to other data mining models. Therefore, it is suggested to researchers in different fields to use DT to predict the studied variables. Also, it is suggested to use other approaches such as random forest or XGBoost to improve the accuracy in future studies. [ABSTRACT FROM AUTHOR]

Background and Aim: Early whitish gastric neoplasms can be easily misdiagnosed; differential diagnosis of gastric whitish lesions remains a challenge. We aim to build a deep learning (DL) model to diagnose whitish gastric neoplasms and explore the effect of adding domain knowledge in model construction. Methods: We collected 4558 images from two institutions to train and test models. We first developed two sole DL models (1 and 2) using supervised and semi‐supervised algorithms. Then we selected diagnosis‐related features through literature research and developed feature‐extraction models to determine features including boundary, surface, roundness, depression, and location. Then predictions of the five feature‐extraction models and sole DL model were combined and inputted into seven machine‐learning (ML) based fitting‐diagnosis models. The optimal model was selected as ENDOANGEL‐WD (whitish‐diagnosis) and compared with endoscopists. Results: Sole DL 2 had higher sensitivity (83.12% vs 68.67%, Bonferroni adjusted P = 0.024) than sole DL 1. Adding domain knowledge, the decision tree performed best among the seven ML models, achieving higher specificity than DL 1 (84.38% vs 72.27%, Bonferroni adjusted P < 0.05) and higher accuracy than DL 2 (80.47%, Bonferroni adjusted P < 0.001) and was selected as ENDOANGEL‐WD. ENDOANGEL‐WD showed better accuracy compared with 10 endoscopists (75.70%, P < 0.001). Conclusions: We developed a novel system ENDOANGEL‐WD combining domain knowledge and traditional DL to detect gastric whitish neoplasms. Adding domain knowledge improved the performance of traditional DL, which provided a novel solution for establishing diagnostic models for other rare diseases potentially. [ABSTRACT FROM AUTHOR]

Background: Temporomandibular disorders (TMD) do not only occur in adults but also in adolescents, with negative impacts on their development. Aim: To propose a predictive model for TMD in adolescents using a decision tree (DT) analysis and to identify groups at high and low risk of developing TMD in the city of Recife, PE, Brazil. Design: This cross‐sectional study was conducted in Recife on 1342 schoolchildren of both sexes aged 10–17 years. The analyses were performed using Pearson's chi‐squared test and Fisher's exact test, as well as the CHAID algorithm for the construction of the DT. The SPSS statistical program was used. Results: The prevalence of TMD was 33.2%. Statistically significant associations were observed between TMD and sex, depression, self‐reported orofacial pain, and orofacial pain on clinical examination. The DT consisted of self‐reported orofacial pain, orofacial pain on physical examination, and depression, with an overall predictive power of 73.0%. Conclusion: The proposed tree has a good predictive capacity and permits to identify groups at high risk of developing TMD among adolescents, such as those with self‐reported orofacial pain or orofacial pain on examination associated with depression. [ABSTRACT FROM AUTHOR]

Machine learning and emergency evacuation prediction share a meaningful association in the context of the Industrial Revolution 4.0 era, given that machine learning can potentially enhance emergency evacuation prediction processes. Machine learning has the potential to revolutionise emergency evacuation prediction by providing a data-driven approach that aids decision-making and response time in emergencies. The complexity of emergency situations can lead to tragic accidents, as individual behaviour heavily influences evacuation time. While some researchers have explored emergency evacuation, existing computer evacuation models lack detailed analysis. To address this gap, this study focuses on reviewing the Random Forest algorithm, an advanced machine learning algorithm based on Decision Trees, in the application of forecasting emergency evacuation time. Through a qualitative research approach using the innovative Hand-searching technique, we conducted a systematic review employing Content Analysis Theory. Multiple sources were examined, including Google Scholar, Science Direct, Scopus, and Universiti Kebangsaan Malaysia e-Journal System. The study's findings shed light on the performance, prediction factors, advantages, and limitations of Random Forest in identifying impacts during an emergency evacuation. These insights hold significant implications for emergency responders, building designers, and policymakers. [ABSTRACT FROM AUTHOR]

Traditionally, the design of an expert system involves acquiring knowledge, in the form of symbolic rules, directly from the expert(s), which is a complex and time‐consuming task. Although expert systems approach is quite old, it is still present, especially where explicit knowledge representation and reasoning, which assure interpretability and explainability, are necessary. Therefore, machine learning methods have been devised to extract rules from data, to facilitate that task. However, those methods are quite inflexible in adapting to the application domain and provide no help in designing the expert system. In this work, we present a framework and corresponding tool, namely ACRES, for semi‐automatically generating expert systems from datasets. ACRES allows for data preprocessing, which helps in structuring knowledge in the form of a tree, called rule hierarchy, which represents (possible) dependencies among data variables and is used for rule formation. This improves interpretability and explainability of the produced systems. We have also designed and evaluated alternative methods for rule extraction from data and for calculation and use of certainty factors, to represent uncertainty; CFs can be dynamically updated. Experimental results on seven well‐known datasets show that the proposed rule extraction methods are comparable to other popular machine learning approaches like decision trees, CART, JRip, PART, Random Forest, and so on, for the classification task. Finally, we give insights on two applications of ACRES. [ABSTRACT FROM AUTHOR]

Background: Patients with Crohn's disease (CD) who lose response to biologics experience reduced quality of life (QoL) and costly hospitalizations. Precision-guided dosing (PGD) provides a comprehensive pharmacokinetic (PK) profile that allows for biologic dosing to be personalized. We analyzed the cost-effectiveness of infliximab (IFX) PGD relative to two other dose intensification strategies (DIS)., Methods: We developed a hybrid (Markov and decision tree) model of patients with CD who had a clinical response to IFX induction. The analysis had a US payer perspective, a base case time horizon of 5 years, and a 4-week cycle length. There were three IFX dosing comparators: PGD; dose intensification based on symptoms, inflammatory markers, and trough IFX concentration (DIS1); and dose intensification based on symptoms alone (DIS2). Patients that failed IFX initiated ustekinumab, followed by vedolizumab, and conventional therapy. Transition probabilities for IFX were estimated from real-world clinical PK data and interventional clinical trial patient-level data. All other transition probabilities were derived from published randomized clinical trials and cost-effectiveness analyses. Utility values were sourced from previous health technology assessments. Direct costs included biologic acquisition and infusion, surgeries and procedures, conventional therapy, and lab testing. The primary outcomes were incremental cost-effectiveness ratios (ICERs). The robustness of results was assessed via one-way sensitivity, scenario, and probabilistic sensitivity analyses (PSA)., Results: PGD was the cost-effective IFX dosing strategy with an ICER of 122,932 $ per quality-adjusted life year (QALY) relative to DIS1 and dominating DIS2. PGD had the lowest percentage (1.1%) of patients requiring a new biologic through 5 years (8.9% and 74.4% for DIS1 and DIS2, respectively). One-way sensitivity analysis demonstrated that the cost-effectiveness of PGD was most sensitive to the time between IFX doses. PSA demonstrated that joint parameter uncertainty had moderate impact on some results., Conclusions: PGD provides clinical and QoL benefits by maintaining remission and avoiding IFX failure; it is the most cost-effective under conservative assumptions., (© 2024 The Authors. Pharmacotherapy: The Journal of Human Pharmacology and Drug Therapy published by Wiley Periodicals LLC on behalf of Pharmacotherapy Publications, Inc.)

This study investigates key performance indicators (KPIs) affecting regional competitiveness in the European Union (EU) and compares its performance across two groups of EU regions identified by their development phase based on the path dependence argument. Utilizing 40 indicators from the Regional Competitiveness Index (RCI) across 268 NUTS2 regions, it identifies five critical KPIs: Knowledge workers, Employment rate, Labor productivity, Potential market size, and not in education, employment, or training. The classification and regression tree decision tree reveals the labor market efficiency pillar as crucial for RCI classification. Offering unique insights into RCI indicators and their impact, this research provides a policy‐oriented perspective and suggests avenues for further comparative research. [ABSTRACT FROM AUTHOR]

Missing values are common, but dealing with them by inappropriate method may lead to large classification errors. Empirical evidences show that the tree‐based classification algorithms such as classification and regression tree (CART) can benefit from imputation, especially multiple imputation. Nevertheless, less attention has been paid to incorporating multiple imputation into cost‐sensitive decision tree induction. This study focuses on the treatment of missing data based on a time‐constrained minimal‐cost tree algorithm. We introduce various approaches to handle incomplete data into the algorithm including complete‐case analysis, missing‐value branch, single imputation, feature acquisition, and multiple imputation. A simulation study under different scenarios examines the predictive performances of the proposed strategies. The simulation results show that the combination of the algorithm with multiple imputation can assure classification accuracy under the budget. A real medical data example provides insights into the problem of missing values in cost‐sensitive learning and the advantages of the proposed methods. [ABSTRACT FROM AUTHOR]

Circular economy (CE) is aimed at redressing the negative impacts on the environment from linear economic models. CE should consider the economic and environmental dimensions; however, in most cases, these have been addressed separately. CE discourses have focused mainly on the environmental dimension and assumed that all material loops must be closed. Despite the calls to engage with CE, there has been extremely limited research on decision‐making for CE. The following steps were followed in the decision‐making for CE framework construction and design: (1) reviewing the literature; (2) considering economic data approaches; (3) accounting for material flows; (4) modifying the fungibility approach to focus on the environmental dimension; (5) considering environmental impacts; and, (6) selecting a decision tree format. This paper combines the economic fungibility (probability of recovery and price), environmental impacts fungibility (probability of recovery and environmental impacts) to develop the 'Decision‐making tree for circular economy' framework. The decision‐making tree for circular economy is aimed at providing guidance on which and how to close loops depending on economic considerations, environmental impact, and probability of recovery of materials. The CE decision‐making tree provides eight options to help decide which and how to close such loops, and help decide which material loops should be eliminated, closed, or remain open. Decision‐making is key to deciding which loops to close to engage with CE more efficiently. [ABSTRACT FROM AUTHOR]

Previous research suggests that group IIA‐secreted phospholipase A2 (sPLA2‐IIA) plays a role in and predicts lethal COVID‐19 disease. The current study reanalyzed a longitudinal proteomic data set to determine the temporal relationship between levels of several members of a family of sPLA2 isoforms and the severity of COVID‐19 in 214 ICU patients. The levels of six secreted PLA2 isoforms, sPLA2‐IIA, sPLA2‐V, sPLA2‐X, sPLA2‐IB, sPLA2‐IIC, and sPLA2‐XVI, increased over the first 7 ICU days in those who succumbed to the disease but attenuated over the same time period in survivors. In contrast, a reversed pattern in sPLA2‐IID and sPLA2‐XIIB levels over 7 days suggests a protective role of these two isoforms. Furthermore, decision tree models demonstrated that sPLA2‐IIA outperformed top‐ranked cytokines and chemokines as a predictor of patient outcome. Taken together, proteomic analysis revealed temporal sPLA2 patterns that reflect the critical roles of sPLA2 isoforms in severe COVID‐19 disease. [ABSTRACT FROM AUTHOR]

Water sustainability in the built environment requires an accurate estimation of residential water end uses (e.g., showers, toilets, faucets, etc.). In this study, we evaluate the performance of four models (Random Forest, RF; Support Vector Machines, SVM; Logistic Regression, Log‐reg; and Neural Networks, NN) for residential water end‐use classification using actual (measured) and synthetic labeled data sets. We generated synthetic labeled data using Conditional Tabular Generative Adversarial Networks. We then utilized grid search to train each model on their respective optimized hyperparameters. The RF model exhibited the best model performance overall, while the Log‐reg model had the shortest execution times under different balanced and imbalanced (based on number of events per class) synthetic data scenarios, demonstrating a computationally efficient alternative for RF for specific end uses. The NN model exhibited high performance with the tradeoff of longer execution times compared to the other classification models. In the balanced data set scenario, all models achieved closely aligned F1‐scores, ranging from 0.83 to 0.90. However, when faced with imbalanced data reflective of actual conditions, both the SVM and Log‐reg models showed inferior performance compared to the RF and NN models. Overall, we concluded that decision tree‐based models emerge as the optimal choice for classification tasks in the context of water end‐use data. Our study advances residential smart water metering systems through creating synthetic labeled end‐use data and providing insight into the strengths and weaknesses of various supervised machine learning classifiers for end‐use identification. Plain Language Summary: We looked at how well different computer models can tell apart types of water use in homes, like identifying when someone is taking a shower versus flushing a toilet. We used real water meter data and also created fake, but realistic, water use data to test these models. Among the models we tested, the Random Forest model (a method that uses a collection of decisions to make predictions) was the most accurate. However, the Logistic Regression model, another type of model we tested, was faster in analyzing the data, making it a good option for quickly identifying specific water uses without needing as much computer power. We also found that all the models we tested were close in detecting what type of water use event had occurred when the data were evenly distributed across different types of water use. But, when the data were uneven—more like real‐life situations—the Random Forest and Neural Network models were better than the others. This research helps improve systems that monitor how humans use water in homes, making it easier to identify where water is used and how we can save more of it, contributing to more sustainable living environments. Key Points: We generated synthetic labeled data representing residential water end uses with a Conditional Tabular Generative Adversarial NetworkIn the context of water end‐use data, decision tree‐based models emerge as the optimal choice for classification tasksLogistic function‐based models, such as logistic regression, are computationally efficient alternatives for classifying specific end uses [ABSTRACT FROM AUTHOR]

Cost‐sensitive ensemble learning as a combination of two approaches, ensemble learning and cost‐sensitive learning, enables generation of cost‐sensitive tree‐based ensemble models using the cost‐sensitive decision tree (CSDT) learning algorithm. In general, tree‐based models characterize nice graphical representation that can explain a model's decision‐making process. However, the depth of the tree and the number of base models in the ensemble can be a limiting factor in comprehending the model's decision for each sample. The CSDT models are widely used in finance (e.g., credit scoring and fraud detection) but lack effective explanation methods. We previously addressed this gap with cost‐sensitive tree Shapley Additive Explanation Method (CSTreeSHAP), a cost‐sensitive tree explanation method for the single‐tree CSDT model. Here, we extend the introduced methodology to cost‐sensitive ensemble models, particularly cost‐sensitive random forest models. The paper details the theoretical foundation and implementation details of CSTreeSHAP for both single CSDT and ensemble models. The usefulness of the proposed method is demonstrated by providing explanations for single and ensemble CSDT models trained on well‐known benchmark credit scoring datasets. Finally, we apply our methodology and analyze the stability of explanations for those models compared to the cost‐insensitive tree‐based models. Our analysis reveals statistically significant differences between SHAP values despite seemingly similar global feature importance plots of the models. This highlights the value of our methodology as a comprehensive tool for explaining CSDT models. [ABSTRACT FROM AUTHOR]

Summary: There are significant developments in the Internet of Vehicles (IoV) field, and the requirements needed in this area are increasing rapidly. When these needs are examined in the near future, it appears that the demand for connected, autonomous, shared, and electric vehicles will increase. Therefore, fundamental problems such as big data flow and storage will arise in the IoV field. Another problem is the delay sensitivity of IoVs and the need to minimize data loss. The use of edge computing (EC) tools can play an important role in obtaining effective solutions to overcome these problems. Delay, bandwidth, and energy consumption rate, which are important qualities in EC systems, emerge as a problem that needs to be improved for delay‐sensitive systems. These improvements belong to the category of nonlinear challenging problems. Effective optimization or machine learning methods can be used to improve these types of problems. In this study, a two‐stage machine learning method is proposed for a more efficient task completion rate and service time. According to the proposed method, in the first stage, the decision tree algorithm is used to select the computing tool to which the task will be sent, and the decision is made on which computing tool to send it to. In the second stage, a linear regression‐based classification method is used to select a delay‐sensitive computing tool. The performance analysis of the proposed method was made using the edgeCloudSim simulation tool, and according to the results obtained, the proposed method provides better results than other algorithms in the literature. [ABSTRACT FROM AUTHOR]

Geographic and seasonal effects are important in driving selection decisions in rice breeding research. Adopting new strategies for characterizing environmental–phenotype associations is critical to understanding these effects, and the outcomes of their study could reflect the benefits of developing locally adapted cultivars. This study aimed to characterize Brazil's tropical irrigated rice (IR) environment, Latin America's largest rice production system. We integrated unsupervised (K‐means clustering) and supervised (decision tree classifier) algorithms to identify environmental clusters (EC) based on historical yield data. The data set included 31 locations and 471 genotypes from 1982 to 2017. We used environmental features (EF), such as weather and geography, as input variables for our analysis, assuming the model as EC ∼ f (EF). Results indicate that the tropical IR production region can be divided into four primary breeding zones, with temperature emerging as a significant factor in the study area. After employing a linear mixed model analysis, we observed that the current relationship between genetics (G), environmental variation (E), and their interaction (G×E) in Brazil's tropical IR has a 1:6:2 ratio. However, when introducing our data‐driven model based on EC, we reduced this ratio to 1:5:1. Therefore, the selection for local adaptability across a large region became more reliable. Our approach successfully identified EC in Brazil's tropical production region of IR, providing valuable insights for defining breeding zones and identifying more productive and stable seed production fields. Core Ideas: Machine learning links phenotypic and environmental features to cluster trialing locations for rice breeding.This approach also revealed the climatic causes of environmental quality for a group of genotypes or cultivars.As an alternative approach to reducing costs, the strategy benefits many stages of breeding and postbreeding.This approach facilitates postbreeding (cultivar recommendations and optimization of seed production sites). [ABSTRACT FROM AUTHOR]

Increased imports of plants and timber through global trade networks provide frequent opportunities for the introduction of novel plant pathogens that can cross‐over from commercial to natural environments, threatening native species and ecosystem functioning. Prevention or management of such outbreaks relies on a diversity of cross‐sectoral stakeholders acting along the invasion pathway. Yet, guidelines are often only produced for a small number of stakeholders, missing opportunities to consider ways to control outbreaks in other parts of the pathway. We used the infection of common juniper with the invasive pathogen Phytophthora austrocedri as a case study to explore the utility of decision tools for managing outbreaks of plant pathogens in the wider environment. We invited stakeholders who manage or monitor juniper populations or supply plants or management advice to participate in a survey exploring their awareness of, and ability to use, an existing decision tree produced by a coalition of statutory agencies augmented with new distribution maps designed by the authors. Awareness of the decision tree was low across all stakeholder groups including those planting juniper for restoration purposes. Stakeholders requested that decision tools contain greater detail about environmental conditions that increase host vulnerability to the pathogen, and clearer examples of when management practices implicated in pathogen introduction or spread should not be adopted. The results demonstrate the need to set clear objectives for the purpose of decision tools and to frame and co‐produce them with many different stakeholders, including overlooked groups, such as growers and advisory agents, to improve management of pathogens in the wider environment. [ABSTRACT FROM AUTHOR]

Background: Lung cancer is the leading cause of cancer‐related death worldwide. This study aimed to establish novel multiclassification prediction models based on machine learning (ML) to predict the probability of malignancy in pulmonary nodules (PNs) and to compare with three published models. Methods: Nine hundred fourteen patients with PNs were collected from four medical institutions (A, B, C and D), which were organized into tables containing clinical features, radiologic features and laboratory test features. Patients were divided into benign lesion (BL), precursor lesion (PL) and malignant lesion (ML) groups according to pathological diagnosis. Approximately 80% of patients in A (total/male: 632/269, age: 57.73 ± 11.06) were randomly selected as a training set; the remaining 20% were used as an internal test set; and the patients in B (total/male: 94/53, age: 60.04 ± 11.22), C (total/male: 94/47, age: 59.30 ± 9.86) and D (total/male: 94/61, age: 62.0 ± 11.09) were used as an external validation set. Logical regression (LR), decision tree (DT), random forest (RF) and support vector machine (SVM) were used to establish prediction models. Finally, the Mayo model, Peking University People's Hospital (PKUPH) model and Brock model were externally validated in our patients. Results: The AUC values of RF model for MLs, PLs and BLs were 0.80 (95% CI: 0.73–0.88), 0.90 (95% CI: 0.82–0.99) and 0.75 (95% CI: 0.67–0.88), respectively. The weighted average AUC value of the RF model for the external validation set was 0.71 (95% CI: 0.67–0.73), and its AUC values for MLs, PLs and BLs were 0.71 (95% CI: 0.68–0.79), 0.98 (95% CI: 0.88–1.07) and 0.68 (95% CI: 0.61–0.74), respectively. The AUC values of the Mayo model, PKUPH model and Brock model were 0.68 (95% CI: 0.62–0.74), 0.64 (95% CI: 0.58–0.70) and 0.57 (95% CI: 0.49–0.65), respectively. Conclusions: The RF model performed best, and its predictive performance was better than that of the three published models, which may provide a new noninvasive method for the risk assessment of PNs. [ABSTRACT FROM AUTHOR]

Supply chain risk management is crucial for the management of sports goods companies, which can help enterprises better complete the transparent and efficient management of the whole process from production to sales. However, there are inevitably some potential risks. Thereby, to study the supply chain risk management of badminton supplies companies in the general environment, the supply chain risk evaluation index system of badminton supplies companies is established in this study by studying the connotation of supply chain risk management and combining the steps of fuzzy comprehensive evaluation (FCE) method. Then, according to these evaluation indexes, the modelling process of the classification and regression tree (CART) algorithm in the corporate overall supply chain risk decision tree is proposed. Moreover, the corporate overall supply chain risk decision tree is established. Moreover, the financial data of 10 A‐share listed companies of badminton and other sports goods listed on the Shanghai Stock Exchange in 2020 are selected as the sample dataset for model training and testing. The potential risks of badminton enterprises are predicted. The results show that the scale of badminton enterprises has been increasing gradually since 2018, but the return on total assets has steadily decreased with the increase of years. It shows that the badminton supplies enterprise's input and output assets are unbalanced, and the overall asset operation effect is relatively poor. However, the growth rate of total assets shows a significant upward trend from 2018 to 2020, rising from 4% to 12.69%. It indicates that the corporate investment and domestic and foreign investment expansion in recent years are high‐speed. The FCE method is employed to obtain the maximum eigenvalue λmax of the enterprise, which is 5.4698. Then, the C.I value is 0.1205, the R.I value is 1.12, and the consistency ratio value is 0.0958 < 0.1, indicating that the judgement matrix is acceptable. It suggests that under the influence of the macro environment and economic situation, the sales situation of the badminton supplies company is subtly affected by the macro environment, and there is a great business risk. [ABSTRACT FROM AUTHOR]

Objective: Body mass index (BMI) is the primary criterion differentiating anorexia nervosa (AN) and atypical anorexia nervosa despite prior literature indicating few differences between disorders. Machine learning (ML) classification provides us an efficient means of accurately distinguishing between two meaningful classes given any number of features. The aim of the present study was to determine if ML algorithms can accurately distinguish AN and atypical AN given an ensemble of features excluding BMI, and if not, if the inclusion of BMI enables ML to accurately classify between the two. Methods: Using an aggregate sample from seven studies consisting of individuals with AN and atypical AN who completed baseline questionnaires (N = 448), we used logistic regression, decision tree, and random forest ML classification models each trained on two datasets, one containing demographic, eating disorder, and comorbid features without BMI, and one retaining all features and BMI. Results: Model performance for all algorithms trained with BMI as a feature was deemed acceptable (mean accuracy = 74.98%, mean area under the receiving operating characteristics curve [AUC] = 74.75%), whereas model performance diminished without BMI (mean accuracy = 59.37%, mean AUC = 59.98%). Discussion: Model performance was acceptable, but not strong, if BMI was included as a feature; no other features meaningfully improved classification. When BMI was excluded, ML algorithms performed poorly at classifying cases of AN and atypical AN when considering other demographic and clinical characteristics. Results suggest a reconceptualization of atypical AN should be considered. Public Significance: There is a growing debate about the differences between anorexia nervosa and atypical anorexia nervosa as their diagnostic differentiation relies on BMI despite being similar otherwise. We aimed to see if machine learning could distinguish between the two disorders and found accurate classification only if BMI was used as a feature. This finding calls into question the need to differentiate between the two disorders. [ABSTRACT FROM AUTHOR]