This work presents a novel method for applying test-time augmentation (TTA) to tabular data. We used TTA along with an ensemble of 42 models to achieve higher performance on the MIT Global Open Source Severity of Illness Score dataset consisting of 131,051 ICU visits and outcomes. This method achieved an AUC of 0.915 on the private test set (19,669 admissions) and won first place at Stanford University’s WiDS Datathon 2020 challenge on Kaggle, while the Acute Physiology and Chronic Health Evaluation (APACHE) IV model (commonly used for ICU survival prediction in the literature) achieved an AUC of 0.868. In addition to increasing the AUC score, our method also reduces “unfair” bias.
Excessive drinking is a major risk factor that leads to many health complications. The diagnosis of alcoholism is challenging, especially when the standard diagnostic tests rely on blood tests and questionnaires that are subjective to the patient and the examiner. The study’s major goal is to find new EEG classification methods to improve past findings and construct a robust EEG classification algorithm to generate accurate predictions with explainable results. The EEG records were examined from two different perspectives and combined with an ensemble of classification models. The first approach was temporal data, and the second was images derived from the original signals. Using fast Fourier transform (FFT) and independent component analysis (ICA), we convert 64-channel temporal data into images along with applying the Symbolic Aggregate approXimation (SAX) technique. Our model combines input data in tabular, temporal, and image formats with an ensemble of linear neural networks, long short-term memory (LSTM), and efficient-net classification models. We have evaluated our method using a publicly available dataset for EEG classification of alcoholic and non-alcoholic subjects. Overall, our algorithm’s highest cross-validation classification accuracy is 85.52% compared to the state-of-the-art EEG-NET’s accuracy of 81.19%.
Machine learning-based Network Intrusion Detection Systems (NIDS) are designed to protect networks by identifying anomalous behaviors or improper uses. In recent years, advanced attacks, such as those mimicking legitimate traffic, have been developed to avoid alerting such systems. Previous works mainly focused on improving the anomaly detector itself, whereas in this paper, we introduce a novel method, Test-Time Augmentation for Network Anomaly Detection (TTANAD), which utilizes test-time augmentation to enhance anomaly detection from the data side. TTANAD leverages the temporal characteristics of traffic data and produces temporal test-time augmentations on the monitored traffic data. This method aims to create additional points of view when examining network traffic during inference, making it suitable for a variety of anomaly detector algorithms. Our experimental results demonstrate that TTANAD outperforms the baseline in all benchmark datasets and with all examined anomaly detection algorithms, according to the Area Under the Receiver Operating Characteristic (AUC) metric.
Children with Autism Spectrum Disorder (ASD) often face unique risks during sports activities due to challenges such as motor coordination difficulties, sensory sensitivities, and communication impairments. This paper provides a comprehensive review of the use of wearable sensor technologies to enhance the safety and participation of children with ASD in sports. Utilizing a systematic approach, we analyze 144 papers identified through advanced search methodology. Our findings reveal that wearable sensors can monitor physiological signals like heart rate variability and electrodermal activity and biomechanical signals such as movement patterns to detect early signs of distress, anxiety, or potential injury. The integration of these technologies into sports settings for children with ASD presents significant potential for improving safety, reducing participation barriers, and enhancing overall well-being. Key findings indicate that while the application of wearable sensors in this context is still emerging, early results are promising. However, challenges remain regarding device usability, data privacy, and the need for further research to validate the effectiveness of these technologies in real-world sports environments. This review highlights the importance of interdisciplinary collaboration among researchers, technology developers, educators, and caregivers to develop and implement wearable sensor solutions that are tailored to the unique needs of children with ASD, thereby promoting safer and more inclusive sports participation.
9 June 2025 in IEEE Journal of Biomedical and Health Informatics
Diabetic Retinopathy (DR), a prevalent diabetes complication leading to blindness, often goes undetected until late stages due to patients seeking help only when symptoms manifest and limited experts' availability. To address these challenges, we present a novel temporal integrative machine learning system that harnesses both fundus images and electronic health records (EHR) for early and enhanced DR detection. Our system uniquely processes EHR data by focusing on temporal trends and long-term patient histories, creating thousands of temporal features that capture their evolving dynamics over time and deliver unparalleled model finesse. This dual-model system includes a temporal tabular model that relies solely on historical medical records and a deep learning multi-modal model that combines these records with fundus images. The models were trained and tested using real clinical data from 5,000 patients at Soroka Hospital in Israel, comprising 25,000 retinal images collected over 8 years and electronic health records spanning up to 20 years. Given the primarily unlabeled nature of the data, the training phase employed a pseudo-labeling technique. The models were evaluated and verified by a retina specialist, surpassing existing models with AUROC scores of 0.881 for the temporal-trend EHR model and 0.988 for the multi-modal imaging + EHR model. The integration of historical temporal medical data with imaging offers a more dynamic and comprehensive machine-learning system, enhancing DR detection and offering new insights into associated risk factors. This system not only aids physicians in obtaining a holistic view of a patient's health over time but also facilitates fast identification of individuals at high risk for DR.