Predicting and identifying factors associated with undernutrition among children under five years in Ghana using machine learning algorithms.

Objective:The aim of this study was to develop and validate a machine learning model to predict the risk of comorbid depression in asthma patients.Methods:We conducted a retrospective study of 2464 asthma patients with comorbid depression using National Health and Nutrition Examination Survey (NHANES) data. Feature selection was conducted using the Boruta algorithm and the Least Absolute Shrinkage and Selection Operator (LASSO). Eight machine learning algorithms, namely Decision Tree (DT), k-Nearest Neighbors (KNN), Light Gradient Booster Machine (LGBM), Logistic Regression (LR), Random Forest (RF), Support Vector Machine (SVM), eXtreme Gradient Boosting (XGBoost), and Multilayer Perceptron (MLP), were trained using 5-fold cross-validation methodology. Model performance was evaluated through various metrics such as area under the curve (AUC), accuracy, sensitivity, specificity, F1 score, and decision curve analysis (DCA). Interpretation was conducted using SHapley Additive exPlanations (SHAP) analysis, highlighting feature importance.Results:The training set comprised 1724 participants, while the validation set included 740 participants, with a depression prevalence of 14.45%. Significant predictors identified included hypertension, chronic obstructive pulmonary disease (COPD), stroke, sleep questionnaire (SLQ) scores, smoking status, Poverty Index Ratio (PIR), and educational level. The XGBoost model demonstrated superior performance compared with alternative machine learning (ML) algorithms, achieving an AUC of 0.750, an accuracy of 69.1%, a sensitivity of 68.2%, a specificity of 73.8%, and an F1 score of 79%. The SHAP method identified SLQ, PIR, and education level as the primary decision factors influencing the ML model’s predictions.Conclusion:The XGBoost model effectively predicts the risk of depression in asthma patients, serving as a valuable reference for early clinical identification and intervention.

e16179 Background: To evaluate the predictive potential of preprocedural clinical factors and CT radiomic features for treatment response in hepatocellular carcinoma (HCC) patients after selective internal radiation therapy (SIRT) with yttrium-90 resin microspheres. Methods: A retrospective analysis was conducted on 112 HCC patients treated with SIRT at Tsinghua Chang Gung Hospital between September 2022 and June 2024, with at least three months of follow-up. Patients were divided into a training set (78) and a validation set (34) based on treatment time. Treatment response was assessed using the modified Response Evaluation in Solid Tumors (mRECIST) criteria, classifying complete or partial remission as objective response (OR) and stable or progressive disease as no response (NR). Clinical, laboratory, and radiomic data were collected. Radiomic features were extracted from arterial and portal-phase contrast-enhanced CT scans within two months pre-SIRT, normalized using Z-scores, and redundant features were removed via intraclass correlation coefficients and correlation analysis. Key features were selected through univariate logistic regression, least absolute shrinkage and selection operator (LASSO), variance inflation factor analysis, and stepwise regression. Using these features, a nomogram model was constructed with traditional logistic regression, alongside machine learning models, including logistic regression (LR), naive bayes (NB), support vector machine (SVM), random forest (RF), extreme gradient boosting (XGBoost), and deep neural networks (DNN). Model performance was evaluated by the area under the curve (AUC), and ROC curves were compared using the DeLong test. Results: Seven clinical features were selected. AUCs for the Nomogram, LR, NB, SVM, RF, XGBoost, and DNN models were 0.919, 0.915, 0.900, 0.953, 0.985, 0.979, and 0.901 in the training set, and 0.760, 0.760, 0.792, 0.757, 0.740, 0.802, and 0.774 in the validation set. Six radiomics features were selected, yielding training set AUCs of 0.927, 0.929, 0.888, 0.946, 1.000, 1.000, and 0.986, and validation set AUCs of 0.681, 0.670, 0.568, 0.611, 0.576, 0.608, and 0.663. In the combined analysis (5 radiomic and 2 clinical features), the AUCs were 0.943, 0.943, 0.914, 0.931, 1.000, 1.000, and 0.959 in the training set, and 0.736, 0.726, 0.623, 0.660, 0.679, 0.670, and 0.646 in the validation set. Machine learning models, particularly RF and XGBoost, outperformed traditional statistical models in the training set (p < 0.05), though no significant differences were observed in the validation set, where traditional models remained robust. Conclusions: Models integrating clinical and radiomic features, developed using statistical and machine learning algorithms, show promise for predicting response to SIRT in HCC patients.

BackgroundPrediction models with high accuracy rates for nonmetastatic cervical cancer (CC) patients are limited. This study aimed to construct and compare predictive models on the basis of machine learning (ML) algorithms for predicting the 5‐year survival status of CC patients through using the Surveillance, Epidemiology, and End Results public database of the National Cancer Institute.MethodsThe data registered from 2004 to 2016 were extracted and randomly divided into training and validation cohorts (8:2). The least absolute shrinkage and selection operator (LASSO) regression was employed to identify significant factors. Then, four predictive models were constructed, including logistic regression (LR), random forest (RF), support vector machine (SVM), and extreme gradient boosting (XGBoost). The predictive models were evaluated and compared using Receiver‐operating characteristics with areas under the curves (AUCs) and decision curve analysis (DCA), respectively.ResultsA total of 13,802 patients were involved and classified into training (N = 11,041) and validation (N = 2761) cohorts. By using the LASSO regression method, seven factors were identified. In the training cohort, the XGBoost model showed the best performance (AUC = 0.8400) compared to the other three models (all p < 0.05 by Delong's test). In the validation cohort, the XGBoost model also demonstrated a superior prediction ability (AUC = 0.8365) than LR and SVM models (both p < 0.05 by Delong's test), although the difference was not statistically significant between the XGBoost and the RF models (p = 0.4251 by Delong's test). Based on the DCA results, the XGBoost model was also superior, and feature importance analysis indicated that the tumor stage was the most important variable among the seven factors.ConclusionsThe XGBoost model proved to be an effective algorithm with better prediction abilities. This model is proposed to support better decision‐making for nonmetastatic CC patients in the future.

Background/Objectives: Hospital readmissions are a key quality metric impacting both patient outcomes and healthcare costs. Traditional logistic regression models, including the LACE index (Length of stay, Admission type, Comorbidity index, and recent Emergency department visits), are commonly used for readmission risk stratification, though their accuracy may be limited by non-linear interactions with other clinical variables. This study compared the predictive performance of non-linear machine learning (ML) models with stepwise logistic regression (LR) and the LACE index for predicting 30-day general medicine readmissions. Methods: We retrospectively analysed adult general medical admissions at a tertiary hospital in Australia from 1 July 2022 to 30 June 2023. Thirty-two variables were extracted from electronic medical records, including demographics, comorbidities, prior healthcare use, socioeconomic status (SES), laboratory data, and frailty (measured by the Hospital Frailty Risk Score). Predictive models included stepwise LR and four ML algorithms: Least Absolute Shrinkage and Selection Operator (LASSO), random forest, Extreme Gradient Boosting (XGBoost), and artificial neural networks (ANNs). Performance was assessed using the area under the curve (AUC), with comparisons made using DeLong's test. Results: Of 5371 admissions, 1024 (19.1%) resulted in 30-day readmissions. Readmitted patients were older and frailer and had more comorbidities and lower SES. Logistic regression (LR) identified the key predictors of outcomes, including heart failure, alcoholism, nursing home residency, and prior admissions, achieving an AUC of 0.62. LR's performance was comparable to that of the LACE index (AUC = 0.61) and machine learning models: LASSO (AUC = 0.63), random forest (AUC = 0.60), and artificial neural networks (ANNs) (AUC = 0.60) (p > 0.05). However, LR significantly outperformed XGBoost (AUC = 0.55) (p < 0.05). Conclusions: About one in five general medicine patients are readmitted within 30 days. Traditional LR performed as well as or better than ML models for readmission risk prediction.

IntroductionLong COVID significantly affects patients' quality of life, yet no standardized treatment has been established. Traditional Chinese Medicine (TCM) presents a promising potential approach with targeted therapeutic strategies. This study aims to develop an explainable machine learning (ML) model and nomogram to identify Long COVID patients who may benefit from TCM, enhancing clinical decision-making.MethodsWe analyzed data from 1,331 Long COVID patients treated with TCM between December 2022 and February 2024 at three hospitals in Zhejiang, China. Effectiveness was defined as improvement in two or more symptoms or a minimum 2-point increase in the Traditional Chinese Medicine Syndrome Score (TCMSS). Data included 11 patient and disease characteristics, 18 clinical symptoms and syndrome scores, and 12 auxiliary examination indicators. The least absolute shrinkage and selection operator (LASSO) method identified features linked to TCM efficacy. Data from 1,204 patients served as the training set, while 127 patients formed the testing set.ResultsWe employed five ML algorithms: Support Vector Machine (SVM), Random Forest (RF), K-Nearest Neighbors (KNN), Extreme Gradient Boosting (XGBoost), and Neural Network (NN). The XGBoost model achieved an Area Under the Curve (AUC) of 0.9957 and an F1 score of 0.9852 in the training set, demonstrating superior performance in the testing set with an AUC of 0.9059 and F1 score of 0.9027. Key features identified through SHapley Additive exPlanations (SHAP) included chest tightness, aversion to cold, age, TCMSS, Short Form (36) Health Survey (SF-36), C-reactive protein (CRP), and lymphocyte ratio. The logistic regression-based nomogram demonstrated an AUC of 0.9479 and F1 score of 0.9384 in the testing set.ConclusionThis study utilized multicenter data and multiple ML algorithms to create a ML model for predicting TCM efficacy in Long COVID treatment. Furthermore, a logistic regression-based nomogram was developed to assist the model and improve decision-making efficiency in TCM applications for Long COVID management.

This study aimed to explore the effectiveness of predicting disease activity in patients with inflammatory bowel disease (IBD), using machine learning (ML) models. A retrospective research was undertaken on IBD patients who were admitted intothe First Affiliated Hospital of Wenzhou Medical University between September 2011 and September 2019. At first, data were randomly split into a 3:1 ratio of training to test set. The least absolute shrinkage and selection operator (LASSO) algorithm was applied to reduce the dimension of variables. These variables were used to generate seven ML algorithms, namely random forests (RFs), adaptive boosting (AdaBoost), K-nearest neighbors (KNNs), support vector machines (SVMs), naïve Bayes (NB), ridge regression, and eXtreme gradient boosting (XGBoost) to train to predict disease activity in IBD patients. SHapley Additive exPlanation (SHAP) analysis was performed to rank variable importance. A total of 876 participants with IBD, consisting of 275 ulcerative colitis (UC) and 601 Crohn's disease (CD), were retrospectively enrolled in the study. Thirty-three variables were obtained from the clinical characteristics and laboratory tests of the participants. Finally, after LASSO analysis, 11 and 5 variables were screened out to construct ML models for CD and UC, respectively. All seven ML models performed well in predicting disease activity in the CD and UC test sets. Among these ML models, SVM was more effective in predicting disease activity in the CD group, whose AUC reached 0.975, sensitivity 0.947, specificity 0.920, and accuracy 0.933. AdaBoost performed best for the UC group, with an AUC of 0.911, sensitivity 0.844, specificity 0.875, and accuracy 0.855. ML algorithms were available and capable of predicting disease activity in IBD patients. Based on clinical and laboratory variables, ML algorithms demonstrate great promise in guiding physicians' decision-making.

ObjectiveAnalyze the psychological and clinical factors of clinically significant tinnitus (THI score ≥38) in patients with hearing loss, construct predictive models based on four machine learning (ML) algorithms, and compare the predictive performance of different models.MethodsPatients with hearing loss who visited the Department of Otolaryngology at Qinghai University between August 2024 and May 2025 were enrolled in this study. Clinical data were retrieved from the hospital’s electronic medical record system. The study outcome was the occurrence of clinically significant tinnitus. Predictive variables were screened using univariate analysis, the least absolute shrinkage and selection operator (LASSO) regression, and the Boruta algorithm. Four ML algorithms—logistic regression (LR), random forest (RF), extreme gradient boosting (XGBoost), and support vector machine (SVM)—were applied to construct and validate predictive models. The area under the receiver operating characteristic curve (AUC) of each model in the validation set was compared using the DeLong test. Additionally, model performance metrics in the validation set were compared to identify the optimal model. Finally, the Shapley additive explanations (SHAP) algorithm was employed to interpret the best-performing model.ResultsNine key variables—age, hypertension, sleep disorder, anxiety, hearing loss severity, depression, noise exposure history, hearing side, and ototoxic drug use—were retained after LASSO and Boruta feature selection. Among the four ML models, the RF algorithm achieved the best predictive performance, with an AUC of 0.973 in the training set and 0.977 in the validation set, followed by XGBoost (AUC = 0.962 and 0.961, respectively). DeLong tests confirmed that RF significantly outperformed LR and SVM models (p < 0.001), while its difference from XGBoost was not significant. In the validation set, the RF model yielded the highest accuracy (0.923), sensitivity (0.929), specificity (0.914), precision (0.945), and F1-score (0.937). SHAP analysis indicated that hearing loss severity, age, and sleep disorder were the most influential predictors, suggesting that both auditory and non-auditory factors contribute substantially to the risk of clinically significant tinnitus.ConclusionThe RF model showed the best performance in predicting clinically significant tinnitus, with hearing loss severity, age, and sleep disorder identified as major predictors. Integrating auditory and psychological factors can improve early risk identification in patients with hearing loss.

Developing and validating a machine learning model to predict chronic pain following total knee arthroplasty.

BackgroundMetabolic dysfunction‐associated steatotic liver disease (MASLD) is an emerging global health concern. This study was aimed at exploring the feasibility of utilizing machine learning (ML) algorithms to predict MASLD in large general populations based on simple anthropometric and biochemical parameters.MethodsData from the 2017–2020 cycles of the US National Health and Nutrition Examination Survey (NHANES) were utilized. A total of 6814 participants (53.0% female) with complete transient elastography data were included. MASLD was defined as a controlled attenuation parameter ≥ 280 dB/m, with cardiometabolic risk factor and without excessive alcohol use. Key characteristics and biomarkers associated with MASLD were identified using the least absolute shrinkage and selection operator (LASSO) and the Boruta algorithms. ML methods, including logistic regression (LR), extreme gradient boosting (XGBoost), bootstrap aggregating, random forest, naive Bayes, light gradient boosting machine (LightGBM), decision tree, and support vector machines, were employed to develop the MASLD prediction models.ResultsThe median age of the 6814 participants was 53 years (interquartile range: 37~65). MASLD was detected among 2611 (38.3%) participants. Key predictors selected via LASSO and Boruta algorithms included body weight, standing height, waist circumference, diagnosis of diabetes, alanine aminotransferase, aspartate aminotransferase, and gamma glutamyl transferase. The areas under the receiver operating characteristic curves of LR, XGBoost, and other ML models were 0.841, 0.837, 0.815, 0.838, 0.814, 0.842, 0.796, and 0.828 in the internal validation cohort. Results indicate that LR, XGBoost, and LightGBM models outperform other models in predicting MASLD.ConclusionsThe ML models of LR, XGBoost, and LightGBM are effective and simplified tools for predicting MASLD in the US general population. This study underscores the potential of ML models with simple noninvasive biomarkers in enhancing early detection and personalized management of fatty liver disease.

BackgroundPrediabetes is an intermediate stage between normal glucose metabolism and diabetes and is associated with increased risk of complications like cardiovascular disease and kidney failure.ObjectiveIt is crucial to recognize individuals with prediabetes early in order to apply timely intervention strategies to decelerate or prohibit diabetes development. This study aims to compare the effectiveness of machine learning (ML) algorithms in predicting prediabetes and identifying its key clinical predictors.MethodsMultiple ML models are evaluated in this study, including random forest, extreme gradient boosting (XGBoost), support vector machine (SVM), and k-nearest neighbors (KNNs), on a dataset of 4743 individuals. For improved performance and interpretability, key clinical features were selected using LASSO (Least Absolute Shrinkage and Selection Operator) regression and principal component analysis (PCA). To optimize model accuracy and reduce overfitting, we used hyperparameter tuning with RandomizedSearchCV for XGBoost and random forest, and GridSearchCV for SVM and KNN. SHAP (Shapley Additive Explanations) was used to assess model-agnostic feature importance. To resolve data imbalance, SMOTE (Synthetic Minority Oversampling Technique) was applied to ensure reliable classifications.ResultsA cross-validated ROC-AUC (receiver operating characteristic area under the curve) score of 0.9117 highlighted the robustness of random forest in generalizing across datasets among the models tested. XGBoost followed closely, providing balanced accuracy in distinguishing between normal and prediabetic cases. While SVMs and KNNs performed adequately as baseline models, they exhibited limitations in sensitivity. The SHAP analysis indicated that BMI, age, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol emerged as the key predictors across models. The performance was significantly enhanced through hyperparameter tuning; for example, the ROC-AUC for SVM increased from 0.813 (default) to 0.863 (tuned). PCA kept 12 components while maintaining 95% of the variance in the dataset.ConclusionsIt is demonstrated in this research that optimized ML models, especially random forest and XGBoost, are effective tools for assessing early prediabetes risk. Combining SHAP analysis with LASSO and PCA enhances transparency, supporting their integration in real-time clinical decision support systems. Future directions include validating these models in diverse clinical settings and integrating additional biomarkers to improve prediction accuracy, offering a promising avenue for early intervention and personalized treatment strategies in preventive health care.

Knee osteoarthritis (KOA) represents a progressive degenerative disorder characterized by the gradual erosion of articular cartilage. This study aimed to develop and validate biomarker-based predictive models for KOA diagnosis using machine learning techniques. Clinical data from 2594 samples were obtained and stratified into training and validation datasets in a 7:3 ratio. Key clinical features were identified through differential analysis between KOA and control groups, combined with least absolute shrinkage and selection operator (LASSO) regression. The SHapley Additive Planning (SHAP) method was employed to rank feature importance quantitatively. Based on these rankings, predictive models were constructed using Logistic Regression (LR), Random Forest (RF), eXtreme Gradient Boosting (xGBoost), Naive Bayes (NB), Support Vector Machine (SVM), and Decision Tree (DT) algorithms. Models were developed for subsets of variables, including the top 5, top 10, top 15, and all identified features. Receiver operating characteristic (ROC) curves were applied to compare diagnostic performance across models. Additionally, a risk stratification framework for KOA prediction was designed using recursive partitioning analysis (RPA). Using difference analysis and LASSO, 44 critical clinical features were identified. Among these, age, plasma prothrombin time, gender, body mass index (BMI), and prothrombin time and international normalized ratio (PTINR) emerged as the top five features, with SHAP values of 0.1990, 0.0981, 0.0471, 0.0433, and 0.0422, respectively. Machine learning analysis demonstrated that these variables provided robust diagnostic performance for KOA. In the training set, area under the curve (AUC) values for LR, RF, xGBoost, NB, SVM, and DT models were 0.947, 0.961, 0.892, 0.952, 0.885, and 0.779, respectively. Similarly, in the validation dataset, these models achieved AUC values of 0.961, 0.943, 0.789, 0.957, 0.824, and 0.76. Among them, RF consistently exhibited superior diagnostic accuracy for KOA. Additionally, RPA analysis indicated a higher prevalence of KOA among individuals aged 54 years and older. The integration of the top five clinical variables significantly enhanced the diagnostic accuracy for KOA, particularly when employing the RF model. Moreover, the RPA model offered valuable insights to assist clinicians in refining prognostic assessments and optimizing clinical decision-making processes.

e13536 Background: The use of artificial intelligence in breast cancer diagnosis has had a significant impact. Alternative to other tests, machine learning (ML) models are low-cost tools that can establish an objective prognosis by considering all the clinical, genomic, and histological data from patients. To evaluate the performance of 12 ML models in predicting the overall survival of breast cancer patients at 60 months of follow-up. Methods: Clinical data was obtained from “The Cancer Genome Atlas-Breast Cancer" database. Preprocessing of data ruled out variables with insufficient values or poor relevance. Data was divided into an 80/20 ratio to perform the training and testing of the models. The models used were: Logistic Regression (LR), Ridge Classifier (RC), Least Absolute Shrinkage and Selection Operator (LASSO), K-nearest Neighbors (KNN), Naive Bayes (NB), Linear Discriminant Analysis (LDA), Decision Tree (DT), Multilayer Perceptron (MP), Stochastic Gradient Descent (SGD), Support Vector Machine (SVM), Random Forest (RF), and XGBoost (XGB). Initial training used the default hyperparameters of each model, performance was evaluated through a 5-fold-cross-validation, the results were used with the GridSearchCV tool to optimize the hyperparameters for the final training and testing of the models. Results: The models with most accuracy were NB (86.76%) and SGD (85.29%), with most sensitivity were: SGD (93.75%), MP (91.67%) and NB (89.89%), with the most specificity were: KNN (95.0%), DT (90.0%) and LASSO (85.0%), with the highest area under the ROC curve were: LDA (90.83%), LASSO (90.1%), and LR (89.79%). The variables with the highest relevance were: "Previous diagnosis of cancer", "Presence of tumor", "Ancillary Therapy", and "Histology". Integrating the best-performing models into an interactive tool preserves the best features without sacrificing efficiency. Conclusions: The models with the best overall performance in predicting the prognosis of patients with breast cancer were NB and LASSO. The best way to evaluate the performance of a predictive prognosis tool is the area under the ROC curve. For developing countries, an AI tool to predict the patient outcome is viable when other expensive prognosis genomic tools are not available. Further studies with data from developing countries are needed to improve the performance of this tool.

IntroductionAcute ischemic stroke (AIS) patients often experience poor functional outcomes post-intravenous thrombolysis (IVT). Novel computational methods leveraging machine learning (ML) architectures increasingly support medical decision-making. We aimed to develop and validate a machine learning model to predict 3-month unfavorable functional outcome after IVT in AIS patients.MethodsThis retrospective study developed ML prognostic models for 3-month functional outcome (modified Rankin scale scores of 3–6) in IVT-treated AIS patients. A derivation cohort (n = 938) was split 7:3 for training/testing, with an independent external validation cohort (n = 324). The least absolute shrinkage and selection operator (LASSO) regression selected predictors from clinical/neuroimaging/laboratory variables. Eight ML algorithms (including Logistic Regression, Random Forest, Extreme Gradient Boosting, Multilayer Perceptron, Support Vector Machine, Light Gradient Boosting Machine, Decision Tree, and K-Nearest Neighbors) were trained using 10-fold cross-validation and evaluated on test/external sets via the area under the curve (AUC), accuracy, precision, recall and F1-score. Additionally, the SHapley Additive exPlanations (SHAP) interpreted the optimal model.Results938 patients constituted the derivation cohort (training: n = 656, test: n = 282) and 324 patients the external validation cohort. Unfavorable 3-month outcomes (mRS 3–6) occurred in 25.7% and 22.8%, respectively. LASSO regression selected five predictors: the neutrophil-to-lymphocyte ratio (NLR), admission National Institutes of Health Stroke Scale (NIHSS) score, the Alberta Stroke Program Early CT Score (ASPECTS), atrial fibrillation, and blood glucose. While tree-based methods like XGBoost and LightGBM showed elevated training performance (e.g., XGBoost training AUC = 0.878) but significant drops in validation (AUC = 0.791), LR demonstrated optimal performance: robust training AUC (0.792), minimal validation degradation (AUC = 0.787). LR model was subsequently employed as classification method demonstrating optimal performance with (AUC = 0.777) in the test dataset. External validation confirmed LR’s stability (AUC = 0.797). SHAP analysis ranked NLR as the strongest predictor (followed by NIHSS/ASPECTS), with higher values increasing risk. Learning curves indicated no overfitting. A nomogram enabled individualized risk quantification.ConclusionA parsimonious 5-variable LR model robustly predicts 3-month post-IVT outcomes, combining clinical utility, interpretability, and generalizability. NLR-driven inflammation is critical to prognosis. This tool facilitates early high-risk patient identification for personalized intervention.

ObjectiveTo build a machine learning (ML) prediction model for prostate cancer (PCa) from transrectal ultrasound video clips of the whole prostate gland, diagnostic performance was compared with magnetic resonance imaging (MRI).MethodsWe systematically collated data from 501 patients—276 with prostate cancer and 225 with benign lesions. From a final selection of 231 patients (118 with prostate cancer and 113 with benign lesions), we randomly chose 170 for the purpose of training and validating a machine learning model, while using the remaining 61 to test a derived model. We extracted 851 features from ultrasound video clips. After dimensionality reduction with the least absolute shrinkage and selection operator (LASSO) regression, 14 features were finally selected and the support vector machine (SVM) and random forest (RF) algorithms were used to establish radiomics models based on those features. In addition, we creatively proposed a machine learning models aided diagnosis algorithm (MLAD) composed of SVM, RF, and radiologists’ diagnosis based on MRI to evaluate the performance of ML models in computer-aided diagnosis (CAD). We evaluated the area under the curve (AUC) as well as the sensitivity, specificity, and precision of the ML models and radiologists’ diagnosis based on MRI by employing receiver operator characteristic curve (ROC) analysis.ResultsThe AUC, sensitivity, specificity, and precision of the SVM in the diagnosis of PCa in the validation set and the test set were 0.78, 63%, 80%; 0.75, 65%, and 67%, respectively. Additionally, the SVM model was found to be superior to senior radiologists’ (SR, more than 10 years of experience) diagnosis based on MRI (AUC, 0.78 vs. 0.75 in the validation set and 0.75 vs. 0.72 in the test set), and the difference was statistically significant (p< 0.05).ConclusionThe prediction model constructed by the ML algorithm has good diagnostic efficiency for prostate cancer. The SVM model’s diagnostic efficiency is superior to that of MRI, as it has a more focused application value. Overall, these prediction models can aid radiologists in making better diagnoses.

Subarachnoid hemorrhage (SAH) denotes a serious type of hemorrhagic stroke that often leads to a poor prognosis and poses a significant socioeconomic burden. Timely assessment of the prognosis of SAH patients is of paramount clinical importance for medical decision making. Currently, clinical prognosis evaluation heavily relies on patients’ clinical information, which suffers from limited accuracy. Non-contrast computed tomography (NCCT) is the primary diagnostic tool for SAH. Radiomics, an emerging technology, involves extracting quantitative radiomics features from medical images to serve as diagnostic markers. However, there is a scarcity of studies exploring the prognostic prediction of SAH using NCCT radiomics features. The objective of this study is to utilize machine learning (ML) algorithms that leverage NCCT radiomics features for the prognostic prediction of SAH. Retrospectively, we collected NCCT and clinical data of SAH patients treated at Beijing Hospital between May 2012 and November 2022. The modified Rankin Scale (mRS) was utilized to assess the prognosis of patients with SAH at the 3-month mark after the SAH event. Based on follow-up data, patients were classified into two groups: good outcome (mRS ≤ 2) and poor outcome (mRS > 2) groups. The region of interest in NCCT images was delineated using 3D Slicer software, and radiomic features were extracted. The most stable and significant radiomic features were identified using the intraclass correlation coefficient, t-test, and least absolute shrinkage and selection operator (LASSO) regression. The data were randomly divided into training and testing cohorts in a 7:3 ratio. Various ML algorithms were utilized to construct predictive models, encompassing logistic regression (LR), support vector machine (SVM), random forest (RF), light gradient boosting machine (LGBM), adaptive boosting (AdaBoost), extreme gradient boosting (XGBoost), and multi-layer perceptron (MLP). Seven prediction models based on radiomic features related to the outcome of SAH patients were constructed using the training cohort. Internal validation was performed using five-fold cross-validation in the entire training cohort. The receiver operating characteristic curve, accuracy, precision, recall, and f-1 score evaluation metrics were employed to assess the performance of the classifier in the overall dataset. Furthermore, decision curve analysis was conducted to evaluate model effectiveness. The study included 105 SAH patients. A comprehensive set of 1316 radiomics characteristics were initially derived, from which 13 distinct features were chosen for the construction of the ML model. Significant differences in age were observed between patients with good and poor outcomes. Among the seven constructed models, model_SVM exhibited optimal outcomes during a five-fold cross-validation assessment, with an average area under the curve (AUC) of 0.98 (standard deviation: 0.01) and 0.88 (standard deviation: 0.08) on the training and testing cohorts, respectively. In the overall dataset, model_SVM achieved an accuracy, precision, recall, f-1 score, and AUC of 0.88, 0.84, 0.87, 0.84, and 0.82, respectively, in the testing cohort. Radiomics features associated with the outcome of SAH patients were successfully obtained, and seven ML models were constructed. Model_SVM exhibited the best predictive performance. The radiomics model has the potential to provide guidance for SAH prognosis prediction and treatment guidance.