wearable-devices

This paper proposes a multimodal framework called PART for predicting injury risk in tennis players using machine learning on data from wearables, questionnaires, and video analysis, showing strong performance in holistic athlete assessment.

This paper investigates reducing the computational complexity of deep neural networks for EEG analysis on wearable devices by applying parameter quantization and electrode reduction techniques, demonstrating significant complexity reduction with minimal accuracy loss for epileptic seizure detection.

VCR is a self-supervised framework that learns robust representations from incomplete wearable signals using orthogonal tokenization and missing-aware mixture-of-experts, improving performance under modality missingness.

Researchers at MIT developed artificial muscles from woven fibers with embedded pumps that use electric fields to move fluid, enabling smooth, quiet motion for soft robotic arms and wearable devices.

This paper introduces LifeDialBench, a novel benchmark for evaluating memory capabilities in continuous lifelog scenarios using wearable devices, and proposes an online evaluation protocol that enforces temporal causality. Key finding: sophisticated memory systems underperform simple RAG baselines, highlighting the importance of high-fidelity context preservation over lossy compression.