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

TL;DR

This study introduces a gait foundation model that uses 3D skeletal motion to predict various health phenotypes, demonstrating gait's potential as an independent systemic biomarker beyond traditional clinical assessments.

Contribution

The paper presents a novel deep learning model trained on large-scale 3D gait data that outperforms traditional features and reveals gait's broad predictive power across multiple health systems.

Findings

Embeddings predict age, BMI, and visceral fat with high accuracy.

Gait provides independent information for 18 body systems.

Legs and torso contribute differently to metabolic and lifestyle predictions.

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.