In this study, we highlighted the growing need for automated electrocardiogram (ECG) signal classification using deep learning to overcome the limitations of traditional ECG interpretation algorithms that can lead to misdiagnosis and inefficiency. Convolutional neural networks (CNN) application to ECG signals is gaining significant attention owing to their exceptional image-classification capabilities. However, we addressed the lack of standardized methods for converting 1D ECG signals into 2D-CNN-compatible input images by using time-frequency methods and selecting hyperparameters associated with these methods, particularly the choice of function. Furthermore, we investigated the effects of fine-tuned training, a technique where pre-trained weights are adapted to a specific dataset, on 2D-CNNs for ECG classification. We conducted the experiments using the MIT-BIH Arrhythmia Database, focusing on classifying premature ventricular contractions (PVCs) and abnormal heartbeats originating from ventricles. We employed several CNN architectures pre-trained on ImageNet and fine-tuned using the proposed ECG datasets. We found that using the Ricker Wavelet function outperformed other feature extraction methods with an accuracy of 96.17%. We provided crucial insights into CNNs for ECG classification, underscoring the significance of fine-tuning and hyperparameter selection in image transformation methods. The findings provide valuable guidance for researchers and practitioners, improving the accuracy and efficiency of ECG analysis using 2D-CNNs. Future research avenues may include advanced visualization techniques and extending CNNs to multiclass classification, expanding their utility in medical diagnosis.

Neuromuscular disorders are diseases that damage our ability to control body movements. Needle electromyography (nEMG) is often used to diagnose neuromuscular disorders, which is an electrophysiological test measuring electric signals generated from a muscle using an invasive needle. Characteristics of nEMG signals are manually analyzed by an electromyographer to diagnose the types of neuromuscular disorders, and this process is highly dependent on the subjective experience of the electromyographer. Contemporary computer-aided methods utilized deep learning image classification models to classify nEMG signals which are not optimized for classifying signals. Additionally, model explainability was not addressed which is crucial in medical applications. This study aims to improve prediction accuracy, inference time, and explain model predictions in nEMG neuromuscular disorder classification. This study introduces the nEMGNet, a one-dimensional convolutional neural network with residual connections designed to extract features from raw signals with higher accuracy and faster speed compared to image classification models from previous works. Next, the divide-and-vote (DiVote) algorithm was designed to integrate each subject's heterogeneous nEMG signal data structures and to utilize muscle subtype information for higher accuracy. Finally, feature visualization was used to identify the causality of nEMGNet diagnosis predictions, to ensure that nEMGNet made predictions on valid features, not artifacts. The proposed method was tested using 376 nEMG signals measured from 57 subjects between June 2015 to July 2020 in Seoul National University Hospital. The results from the three-class classification task demonstrated that nEMGNet's prediction accuracy of nEMG signal segments was 62.35%, and the subject diagnosis prediction accuracy of nEMGNet and the DiVote algorithm was 83.69 %, over 5-fold cross-validation. nEMGNet outperformed all models from previous works on nEMG diagnosis classification, and heuristic analysis of feature visualization results indicate that nEMGNet learned relevant nEMG signal characteristics. This study introduced nEMGNet and DiVote algorithm which demonstrated fast and accurate performance in predicting neuromuscular disorders based on nEMG signals. The proposed method may be applied in medicine to support real-time electrophysiologic diagnosis.

Background: Machine learning (ML) has the potential to enhance performance by capturing nonlinear interactions. However, ML-based models have some limitations in terms of interpretability. Objective: This study aimed to develop and validate a more comprehensible and efficient ML-based scoring system using SHapley Additive exPlanations (SHAP) values. Methods: We developed and validated the Explainable Automated nonlinear Computation scoring system for Health (EACH) framework score. We developed a CatBoost-based prediction model, identified key features, and automatically detected the top 5 steepest slope change points based on SHAP plots. Subsequently, we developed a scoring system (EACH) and normalized the score. Finally, the EACH score was used to predict perioperative stroke. We developed the EACH score using data from the Seoul National University Hospital cohort and validated it using data from the Boramae Medical Center, which was geographically and temporally different from the development set. Results: When applied for perioperative stroke prediction among 38,737 patients undergoing noncardiac surgery, the EACH score achieved an area under the curve (AUC) of 0.829 (95% CI 0.753-0.892). In the external validation, the EACH score demonstrated superior predictive performance with an AUC of 0.784 (95% CI 0.694-0.871) compared with a traditional score (AUC=0.528, 95% CI 0.457-0.619) and another ML-based scoring generator (AUC=0.564, 95% CI 0.516-0.612). Conclusions: The EACH score is a more precise, explainable ML-based risk tool, proven effective in real-world data. The EACH score outperformed traditional scoring system and other prediction models based on different ML techniques in predicting perioperative stroke.

Current risk stratification strategies for patients with hypertrophic cardiomyopathy (HCM) are limited to traditional methodologies. The authors aimed to establish machine learning (ML)-based models to discriminate major cardiovascular events in patients with HCM. We enrolled consecutive HCM patients from 2 tertiary referral centers and used 25 clinical and echocardiographic features to discriminate major adverse cardiovascular events (MACE), including all-cause death, admission for heart failure (HF-adm), and stroke. The best model was selected for each outcome using the area under the receiver operating characteristic curve (AUROC) with 20-fold cross-validation. After testing in the external validation cohort, the relative importance of features in discriminating each outcome was determined using the SHapley Additive exPlanations (SHAP) method. In total, 2,111 patients with HCM (age 61.4 ± 13.6 years; 67.6% men) were analyzed. During the median 4.0 years of follow-up, MACE occurred in 341 patients (16.2%). Among the 4 ML models, the logistic regression model achieved the best AUROC of 0.800 (95% CI: 0.760-0.841) for MACE, 0.789 (95% CI: 0.736-0.841) for all-cause death, 0.798 (95% CI: 0.736-0.860) for HF-adm, and 0.807 (95% CI: 0.754-0.859) for stroke. The discriminant ability of the logistic regression model remained excellent when applied to the external validation cohort for MACE (AUROC = 0.768), all-cause death (AUROC = 0.750), and HF-adm (AUROC = 0.806). The SHAP analysis identified left atrial diameter and hypertension as important variables for all outcomes of interest. The proposed ML models incorporating various phenotypes from patients with HCM accurately discriminated adverse cardiovascular events and provided variables with high importance for each outcome.

Currently, the identification of stroke patients with an increased suicide risk is mainly based on self‐report questionnaires, and this method suffers from a lack of objectivity. This study developed and validated a suicide ideation (SI) prediction model using clinical data and identified SI predictors. Significant variables were selected through traditional statistical analysis based on retrospective data of 385 stroke patients; the data were collected from October 2012 to March 2014. The data were then applied to three boosting models (Xgboost, CatBoost, and LGBM) to identify the comparative and best performing models. Demographic variables that showed significant differences between the two groups were age, onset, type, socioeconomic, and education level. Additionally, functional variables also showed a significant difference with regard to ADL and emotion (p < 0.05). The CatBoost model (0.900) showed higher performance than the other two models; and depression, anxiety, self-efficacy, and rehabilitation motivation were found to have high importance. Negative emotions such as depression and anxiety showed a positive relationship with SI and rehabilitation motivation and self-efficacy displayed an inverse relationship with SI. Machine learning-based SI models could augment SI prevention by helping rehabilitation and medical professionals identify high-risk stroke patients in need of SI prevention intervention.