Freezing of Gait Detection Using Gramian Angular Fields and Federated Learning from Wearable Sensors

2024-11-18

arXiv.org

AI Summary

Real-time freezing-of-gait (FOG) detection in free-living settings for Parkinson’s disease patients remains challenging due to poor cross-subject and temporal generalizability of existing methods, low reliability of multi-sensor setups, privacy-invasive centralized training, and inability to adapt to disease progression. Method: We propose the first single-sensor framework integrating Gramian Angular Field (GAF)-based time-frequency representation with lightweight federated deep learning, operating exclusively on uniaxial waist-mounted accelerometer data and enabling on-device edge inference via smartphones. Contribution/Results: The framework enables privacy-preserving cross-user collaborative optimization, robust modeling under missing data, and individualized adaptive evolution. Experiments demonstrate a 22.2% improvement in F1-score, a 74.53% reduction in false alarm rate, and a 10.4% accuracy gain over uniaxial baselines—significantly enhancing FOG event sensitivity and long-term clinical deployability.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Privacy
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Freezing of gait (FOG) is a debilitating symptom of Parkinson's disease that impairs mobility and safety by increasing the risk of falls. An effective FOG detection system must be accurate, real-time, and deployable in free-living environments to enable timely interventions. However, existing detection methods face challenges due to (1) intra- and inter-patient variability, (2) subject-specific training, (3) using multiple sensors in FOG dominant locations (e.g., ankles) leading to high failure points, (4) centralized, non-adaptive learning frameworks that sacrifice patient privacy and prevent collaborative model refinement across populations and disease progression, and (5) most systems are tested in controlled settings, limiting their real-world applicability for continuous in-home monitoring. Addressing these gaps, we present FOGSense, a real-world deployable FOG detection system designed for uncontrolled, free-living conditions using only a single sensor. FOGSense uses Gramian Angular Field (GAF) transformations and privacy-preserving federated deep learning to capture temporal and spatial gait patterns missed by traditional methods with a low false positive rate. We evaluated our system using a public Parkinson's dataset collected in a free-living environment. FOGSense improves accuracy by 10.4% over a single-axis accelerometer, reduces failure points compared to multi-sensor systems, and demonstrates robustness to missing values. The federated architecture allows personalized model adaptation and efficient smartphone synchronization during off-peak hours, making it effective for long-term monitoring as symptoms evolve. Overall, FOGSense achieved a 22.2% improvement in F1-score and a 74.53% reduction in false positive rate compared to state-of-the-art methods, along with enhanced sensitivity for FOG episode detection.

Problem

• Detect Freezing of Gait accurately in real-world conditions
• Overcome patient variability and privacy issues in detection
• Reduce sensor dependency and improve long-term monitoring

Innovation

• Uses Gramian Angular Field transformations
• Employs federated deep learning
• Single-sensor real-world deployment