Self-Supervised Learning and Opportunistic Inference for Continuous Monitoring of Freezing of Gait in Parkinson's Disease

TL;DR

LIFT-PD tackles the need for practical, in-home FoG monitoring in PD with a label-efficient approach. It combines self-supervised pretraining on unlabeled data, a differential hopping windowing technique to balance training, and an opportunistic inference module to reduce energy use on wearables. The approach achieves precision and accuracy gains over supervised baselines, demonstrates strong generalization across subjects, and shows clinically meaningful correlations with standard PD assessments. With a single waist-worn accelerometer and real-time processing, LIFT-PD offers a scalable, energy-efficient solution for continuous FoG monitoring and timely cueing interventions.

Abstract

Parkinson's disease (PD) is a progressive neurological disorder that impacts the quality of life significantly, making in-home monitoring of motor symptoms such as Freezing of Gait (FoG) critical. However, existing symptom monitoring technologies are power-hungry, rely on extensive amounts of labeled data, and operate in controlled settings. These shortcomings limit real-world deployment of the technology. This work presents LIFT-PD, a computationally-efficient self-supervised learning framework for real-time FoG detection. Our method combines self-supervised pre-training on unlabeled data with a novel differential hopping windowing technique to learn from limited labeled instances. An opportunistic model activation module further minimizes power consumption by selectively activating the deep learning module only during active periods. Extensive experimental results show that LIFT-PD achieves a 7.25% increase in precision and 4.4% improvement in accuracy compared to supervised models while using as low as 40% of the labeled training data used for supervised learning. Additionally, the model activation module reduces inference time by up to 67% compared to continuous inference. LIFT-PD paves the way for practical, energy-efficient, and unobtrusive in-home monitoring of PD patients with minimal labeling requirements.

Figures (11)

• Figure 1: Self-Supervised training pipeline for real-time FoG detection. Pretrain: Encoder reconstructs masked signal segments, outputting predicted values $h_i$ for missing data. Fine-tuning: Encoder weights are frozen while the MLP is optimized using cross-entropy loss on labeled data. End-to-End Pipeline: The trained model is integrated into a real-time system, where model activation module (MAM) selectively activates the computationally intensive FoG detection model for energy-efficient, long-term monitoring.
• Figure 2: Model Activation Module: Activity threshold-based triggering mechanism for computational offloading and battery life extension in wearable FoG detection system.
• Figure 3: DHWT segmentation process for training set (a); FoG proportion using standard vs. DHWT segmentation (b).
• Figure 4: Density distribution and box plot of the Unified Parkinson's Disease Rating Scale (UPDRS) scores, separated by sex. The top panel displays the density distribution of UPDRS scores, with males shown in blue and females in purple. The bottom panel presents a box plot indicating variations in UPDRS scores between sexes, highlighting the median scores, interquartile ranges, and outliers.
• Figure 5: Architecture of stacked 1D CNN model. Input to the model is 3-minute raw sensor data.