A Gait Foundation Model Predicts Multi-System Health Phenotypes from 3D Skeletal Motion

2026-03-26

AI Summary

This study reconceptualizes gait not merely as a symptom of specific diseases but as an independent, multisystem biomarker of overall health. Leveraging 3D skeletal motion data from 3,414 deeply phenotyped adults, the authors develop a foundational gait model using deep learning to embed gait characteristics into a unified representation. Remarkably, without relying on conventional covariates such as age or BMI, the model significantly predicts 1,980 out of 3,210 health phenotypes, achieving strong performance on key metrics including age (r = 0.69), BMI (r = 0.90), and visceral fat area (r = 0.82). It enhances predictive accuracy across 18 and 17 body systems in males and females, respectively, and anatomical ablation experiments elucidate the distinct contributions of different body segments to phenotype prediction.

Technology Category

• Machine Learning: Multimodal Learning
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Gait is increasingly recognized as a vital sign, yet current approaches treat it as a symptom of specific pathologies rather than a systemic biomarker. We developed a gait foundation model for 3D skeletal motion from 3,414 deeply phenotyped adults, recorded via a depth camera during five motor tasks. Learned embeddings outperformed engineered features, predicting age (Pearson r = 0.69), BMI (r = 0.90), and visceral adipose tissue area (r = 0.82). Embeddings significantly predicted 1,980 of 3,210 phenotypic targets; after adjustment for age, BMI, VAT, and height, gait provided independent gains in all 18 body systems in males and 17 of 18 in females, and improved prediction of clinical diagnoses and medication use. Anatomical ablation revealed that legs dominated metabolic and frailty predictions while torso encoded sleep and lifestyle phenotypes. These findings establish gait as an independent multi-system biosignal, motivating translation to consumer-grade video and its integration as a scalable, passive vital sign.

Problem

• gait
• multi-system health
• biomarker
• phenotypic prediction
• 3D skeletal motion

Innovation

• gait foundation model
• 3D skeletal motion
• multi-system health prediction
• deep phenotyping
• anatomical ablation