Tag
This paper proposes a timing-aware, interpretable self-attention framework that encodes wearable behavioral signals as discrete events with onset timestamps and durations, outperforming existing event-based approaches while improving interpretability for task transition detection.
BeatFlow-ECG is a conditional rectified-flow model that reconstructs single-channel ECG from wearable PPG and IMU signals, outperforming deterministic, adversarial, and diffusion baselines on PPG-DaLiA and WESAD with Pearson correlations up to 0.986 and strong R-peak detection F1 scores.
This paper proposes pretraining the CABiGRU temporal architecture on diffusion-generated sensor windows, then fine-tuning on real DEO smartwatch data, to improve recognition of underrepresented eating/drinking activities, achieving 90.6% balanced accuracy. The approach shows that diffusion-based synthetic data better preserves multivariate temporal structure than SMOTE/GAN augmentation for handling class imbalance in human activity recognition.
This paper introduces a wireless smart insole platform and machine-learning workflow for recognizing human activities like sitting, standing, walking, and unstable walking to support continuous monitoring in elderly care, using pressure and inertial signals with high accuracy.
HB-PVI is a hierarchical Bayesian framework that optimizes personalization decisions in complex activity recognition by balancing gains and costs, demonstrating that a population-first deployment policy can reduce labeling expenses while maintaining performance.
This research paper explores the use of 24-hour wrist movement data to study and predict human health conditions and diseases, involving interdisciplinary collaborations across multiple universities.
DeMMO is an interpretable framework for longitudinal, multi-disease, and multi-outcome learning of digital mobility outcomes using multi-task learning, achieving improved prediction performance over baselines on the Mobilise-D dataset.
This paper introduces a two-stage personalized thermal comfort system using reinforcement learning and multimodal sensing to adapt HVAC controls for individual occupant preferences.
A research paper evaluating CHAP, a CNN-BiLSTM model, for classifying sedentary behavior from wearable accelerometer data, focusing on transfer from hip to wrist placement and the benefits of finetuning.
RAG-HAR+ is a retrieval-first, cost-optimized extension of RAG-HAR for human activity recognition from wearable sensors. It uses a retrieval designer agent and majority voting to reduce LLM usage while maintaining accuracy, and demonstrates feasibility for edge deployment.
This paper systematically compares seven state-of-the-art sensor fusion methods for multi-modal human activity recognition on the HARMES dataset, showing that Gated Multi-modal Fusion achieves the highest macro F1-score of 0.82, outperforming the baseline by 6 percentage points.
This paper proposes an interpretable causal-discovery-guided framework for deriving a Sleep Recovery Score (SRS) from multimodal polysomnography data, demonstrating up to 2.5× stronger alignment with perceived recovery than the traditional Apnea–Hypopnea Index (AHI), with potential applications in connected health.
This paper proposes the Personal Care Utility (PCU), an event-driven architecture designed to bridge the gap between episodic clinical care and the 8,759 hours of daily life that shape health. It organizes continuous personal data into meaningful events, uses LLMs for reasoning and communication while grounding clinical decisions in evidence, and instantiates the framework for Type 2 Diabetes management.
AnyMo is a geometry-aware framework for setup-agnostic human motion modeling using physics-grounded IMU simulation and graph encoding, achieving significant improvements in zero-shot activity recognition, cross-modal retrieval, and motion captioning across multiple datasets.