Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

mmWave Radar for Sit-to-Stand Analysis: A Comparative Study with Wearables and Kinect

Shuting Hu, Peggy Ackun, Xiang Zhang, Siyang Cao, Jennifer Barton, Melvin G. Hector, Mindy J. Fain, Nima Toosizadeh

TL;DR

This work investigates a non-contact, privacy-preserving approach to analyze Sit-to-Stand movements using 60 GHz mmWave radar, reconstructing a 17-joint skeleton with the mmPose-FK framework and comparing radar-derived features to Kinect and wearable sensors. The methodology combines radar signal processing (range, Doppler, angle estimation), pose estimation with LOOCV, inverse-kinematics-based joint angles, and cross-sensor synchronization to extract STS features such as duration and ROM. Results show high agreement for STS duration across all sensors and good trunk movement concordance between radar and Kinect, while knee motion proves challenging for radar-based joint-level inference, highlighting strengths and limitations of each modality. The study advocates multi-sensor fusion as a path toward robust, clinic-ready motion analysis for fall risk assessment, and outlines future work in aging populations and more complex tasks like TUG.

Abstract

This study explores a novel approach for analyzing Sit-to-Stand (STS) movements using millimeter-wave (mmWave) radar technology. The goal is to develop a non-contact sensing, privacy-preserving, and all-day operational method for healthcare applications, including fall risk assessment. We used a 60GHz mmWave radar system to collect radar point cloud data, capturing STS motions from 45 participants. By employing a deep learning pose estimation model, we learned the human skeleton from Kinect built-in body tracking and applied Inverse Kinematics (IK) to calculate joint angles, segment STS motions, and extract commonly used features in fall risk assessment. Radar extracted features were then compared with those obtained from Kinect and wearable sensors. The results demonstrated the effectiveness of mmWave radar in capturing general motion patterns and large joint movements (e.g., trunk). Additionally, the study highlights the advantages and disadvantages of individual sensors and suggests the potential of integrated sensor technologies to improve the accuracy and reliability of motion analysis in clinical and biomedical research settings.

mmWave Radar for Sit-to-Stand Analysis: A Comparative Study with Wearables and Kinect

TL;DR

This work investigates a non-contact, privacy-preserving approach to analyze Sit-to-Stand movements using 60 GHz mmWave radar, reconstructing a 17-joint skeleton with the mmPose-FK framework and comparing radar-derived features to Kinect and wearable sensors. The methodology combines radar signal processing (range, Doppler, angle estimation), pose estimation with LOOCV, inverse-kinematics-based joint angles, and cross-sensor synchronization to extract STS features such as duration and ROM. Results show high agreement for STS duration across all sensors and good trunk movement concordance between radar and Kinect, while knee motion proves challenging for radar-based joint-level inference, highlighting strengths and limitations of each modality. The study advocates multi-sensor fusion as a path toward robust, clinic-ready motion analysis for fall risk assessment, and outlines future work in aging populations and more complex tasks like TUG.

Abstract

This study explores a novel approach for analyzing Sit-to-Stand (STS) movements using millimeter-wave (mmWave) radar technology. The goal is to develop a non-contact sensing, privacy-preserving, and all-day operational method for healthcare applications, including fall risk assessment. We used a 60GHz mmWave radar system to collect radar point cloud data, capturing STS motions from 45 participants. By employing a deep learning pose estimation model, we learned the human skeleton from Kinect built-in body tracking and applied Inverse Kinematics (IK) to calculate joint angles, segment STS motions, and extract commonly used features in fall risk assessment. Radar extracted features were then compared with those obtained from Kinect and wearable sensors. The results demonstrated the effectiveness of mmWave radar in capturing general motion patterns and large joint movements (e.g., trunk). Additionally, the study highlights the advantages and disadvantages of individual sensors and suggests the potential of integrated sensor technologies to improve the accuracy and reliability of motion analysis in clinical and biomedical research settings.

Paper Structure

This paper contains 28 sections, 14 equations, 5 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Sit-to-Stand Motion Analysis
  4. Sensor Technologies for STS Analysis
  5. Wearable Sensors
  6. Vision-Based Systems
  7. RF-Based Systems and mmWave Radar
  8. Methodology
  9. Radar Signal Processing
  10. Range Processing
  11. Doppler Processing
  12. Angle Estimation
  13. Pose Estimation
  14. Joint Angle Calculation
  15. Sensors Synchronization
  16. ...and 13 more sections

Figures (5)

  • Figure 1: The proposed workflow of STS analysis using mmWave radar.
  • Figure 2: Recovered STS skeleton with four phases (Sitting: blue, Flexion: orange, Extension: yellow, Standing: purple).
  • Figure 3: Joint angle calculation steps 1 to 3: All joints are positioned relative to the SpineBase. Joint angles are relative values with respect to their parent joints. Five wearable sensors were placed on the waist (W), left and right thigh (T), left and right shank (S).
  • Figure 4: Based on the synchronized signals, segmentation is performed for (a) Kinect, (b) Radar, and (c) Wearable.
  • Figure 5: Bland-Altman plots showing the agreement between different measurement methods.