A sensor-embedded insole system for detecting limb pose and abnormalities through gait analysis

This technology is an insole with built-in force sensors and a processor that analyzes walking patterns to detect limb pose or abnormalities.

Background

Human movement analysis, particularly gait assessment, is a critical field focused on understanding how individuals walk, run, and interact with their environment. Precise evaluation of gait patterns is essential for diagnosing musculoskeletal disorders, preventing injuries, guiding rehabilitation strategies, and enhancing athletic performance. The ability to accurately quantify the forces exerted by the lower limbs during natural movement provides invaluable insights into biomechanical efficiency and potential abnormalities.

Current approaches to gait analysis often present significant limitations. Traditional laboratory setups, utilizing force plates and optical motion capture systems, are expensive, require specialized facilities, and restrict movement to artificial, constrained environments. This makes them unsuitable for continuous monitoring in real-world settings. While some wearable devices offer kinematic data, they frequently lack the detailed, multi-directional force sensing capabilities necessary for comprehensive kinetic analysis. Consequently, obtaining objective, long-term, and ecologically valid force distribution data during everyday activities remains a considerable challenge, hindering personalized assessment and intervention.

Technology

This technology describes a gait analysis system integrated within a foot insole. It features a layered construction with a top and bottom insole pad, designed to be worn by a subject’s limb. Embedded between these pads are multiple force sensors, engineered to detect forces exerted by the limb in at least two directions. A dedicated processor collects the detailed measurement data from these sensors, subsequently analyzing it to determine the limb’s pose or identify any abnormalities by comparing the live data against a predetermined profile.

This technology also includes a method for training an algorithm to accurately predict knee adduction moments based on kinematic and ground reaction force data (US patent 11,963,791). The algorithm is iteratively trained, generating a predictive model useful for: 1. Personalized gait rehabilitation planning 2. Orthopedic implant design optimization 3. Sports injury risk assessment 4. Remote patient mobility monitoring

Benefits

  • Provides detailed measurement of forces exerted by the limb in multiple directions
  • Enables accurate gait analysis through integrated force sensors within the insole
  • Detects abnormalities or irregularities in limb movement or posture
  • Offers real-time data collection and processing for immediate feedback
  • Non-invasive and wearable design suitable for continuous monitoring
  • Supports personalized assessment using predetermined profiles for comparison

Applications

  • Rehabilitation progress monitoring
  • Sports performance optimization
  • Fall risk assessment for elderly
  • Orthopedic disorder diagnosis
  • Custom footwear design

Additional info

Prediction of knee loads during activities of daily living using custom instrumented insoles and machine learning. J Biomech. 2025 Aug 19;191:112921. doi: 10.1016/j.jbiomech.2Snyder SJ, Fakhar M, Miller RH, Bera A, Shim JK. 

Prediction of Medial Tibiofemoral Joint Reaction Force Using Custom Instrumented Insoles and Neural Networks for Walking and Running Tasks. J Appl Biomech. 2025 Jun 23:1-7. doi: 10.1123/jab.20Snyder SJ, Lee H, Chu E, Heo YJ, Miller RH, Shim JK. 

 Prediction of knee adduction moment using innovative instrumented insole and deep learning neural networks in healthy female individuals. Knee. 2023 Mar;41:115-123. doi: 10.1016/j.knee.2022.12.00Snyder SJ, Chu E, Um J, Heo YJ, Miller RH, Shim JK.

US patent 11,963,791 “Training systems and methods for predicting knee adduction moment”

Patent pending 18/797404 “Neural network to predict knee medial joint contact force from custom instrumented insole”

Patent pending 17/778890 “Systems and methods for gait analysis”