More Like this

1. Predicting Lower Extremity Joint Kinematics Using Multi-Modal Data in the Lab and Outdoor Environment

   https://doi.org/10.36227/techrxiv.24302926.v1  

   Hossain, Md_Sanzid Bin; Guo, Zhishan; Sui, Ning; Choi, Hwan (October 2023, 57th Hawaii International Conference on System Sciences (HICSS))

   Predicting future walking joint kinematics is crucial for assistive device control, especially in variable walking environments. Traditional optical motion capture systems provide kinematics data but require laborious post-processing, whereas IMU based systems provide direct calculations but add delays due to data collection and algorithmic processes. Predicting future kinematics helps to compensate for these delays, enabling the system real-time. Furthermore, these predicted kinematics could serve as target trajectories for assistive devices such as exoskeletal robots and lower limb prostheses. However, given the complexity of human mobility and environmental factors, this prediction remains to be challenging. To address this challenge, we propose the Dual-ED-Attention-FAM-Net, a deep learning model utilizing two encoders, two decoders, a temporal attention module, and a feature attention module. Our model outperforms the state-of-the-art LSTM model. Specifically, for Dataset A, using IMUs and a combination of IMUs and videos, RMSE values decrease from 4.45° to 4.22° and from 4.52° to 4.15°, respectively. For Dataset B, IMUs and IMUs combined with pressure insoles result in RMSE reductions from 7.09° to 6.66° and from 7.20° to 6.77°, respectively. Additionally, incorporating other modalities alongside IMUs helps improve the performance of the model. 

   more » « less
2. ULTRA-MoCap: A Multimodal IMU and sEMG Dataset for Upper Body Joint Kinematics Analysis

   https://doi.org/10.6084/m9.figshare.28751156  

   Fritsche, Oliver; Camacho, Steven; Hossain, Md_Sanzid Bin; Halpenny, Tyler; Archniegas, Carlos; Dranetz, Joseph; Hadley, Dexter; Guo, Zhishan; Choi, Hwan (August 2025, Scientific Data - Springer Nature)

   Dataset files related to the paper ULTRA-MoCap: A Multimodal IMU and sEMG Dataset for Upper Body Joint Kinematics AnalysisAuthors:Oliver Fritsche, Steven Camacho, Md Sanzid Bin Hossain, Tyler Halpenny, Carlos Archniegas, Joseph Dranetz, Dexter Hadley, Zhishan Guo, and Hwan ChoiAbstract: Predicting human kinematics from multi-modal sensor data has applications in rehabilitation, sports analysis, assistive device control, and human-computer interaction, offering less intrusive alternatives to camera-based motion capture (MoCap) systems. However, current datasets often lack comprehensive multi-modal data integrating both IMU and sEMG sensors, with limited coverage for upper limb kinematics and insufficient capture of high-fidelity muscle activation and joint dynamics, especially for complex shoulder and elbow movements. These limitations impede the development of robust models for advanced applications. To address these gaps, we introduce the Upper Limb Tracking with Multi-modal Capture (ULTra-MoCap) dataset. It includes IMUs on the hand, wrist, and forearm, and combined sEMG/IMU sensors on the biceps brachii, triceps brachii, and deltoid, enabling detailed tracking of multiple degrees of freedom upper limb kinematics. Data was collected from 13 healthy subjects using Vicon Vero motion capture cameras and Delsys Trigno sEMG/IMU sensors with various upper limb movements. This paper details the data collection, sensor placement, and processing pipeline, establishing ULTRA-MoCap as a benchmark for deep learning applications.________________________________________________________ULTra-MoCap-processed contains already processed data reading for downstream deep learning.ULTra-MoCap-raw contains subjects [1-6](0) and [7-13](1) raw data for reproducibility._________________________________________________________________ 

   more » « less
3. Wearable Sensor-Based Step Length Estimation During Overground Locomotion Using a Deep Convolutional Neural Network

   https://doi.org/10.1109/EMBC46164.2021.9630060  

   Jin, Heejoo; Kang, Inseung; Choi, Gayeon; Molinaro, Dean D.; Young, Aaron J. (November 2021, 2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC))

   Step length is a critical gait parameter that allows a quantitative assessment of gait asymmetry. Gait asymmetry can lead to many potential health threats such as joint degeneration, difficult balance control, and gait inefficiency. Therefore, accurate step length estimation is essential to understand gait asymmetry and provide appropriate clinical interventions or gait training programs. The conventional method for step length measurement relies on using foot-mounted inertial measurement units (IMUs). However, this may not be suitable for real-world applications due to sensor signal drift and the potential obtrusiveness of using distal sensors. To overcome this challenge, we propose a deep convolutional neural network-based step length estimation using only proximal wearable sensors (hip goniometer, trunk IMU, and thigh IMU) capable of generalizing to various walking speeds. To evaluate this approach, we utilized treadmill data collected from sixteen able-bodied subjects at different walking speeds. We tested our optimized model on the overground walking data. Our CNN model estimated the step length with an average mean absolute error of 2.89 ± 0.89 cm across all subjects and walking speeds. Since wearable sensors and CNN models are easily deployable in real-time, our study findings can provide personalized real-time step length monitoring in wearable assistive devices and gait training programs. 

   more » « less
4. A human lower-limb biomechanics and wearable sensors dataset during cyclic and non-cyclic activities

   https://doi.org/10.1038/s41597-023-02840-6  

   Scherpereel, Keaton; Molinaro, Dean; Inan, Omer; Shepherd, Max; Young, Aaron (December 2023, Scientific Data)

   Abstract Tasks of daily living are often sporadic, highly variable, and asymmetric. Analyzing these real-world non-cyclic activities is integral for expanding the applicability of exoskeletons, protheses, wearable sensing, and activity classification to real life, and could provide new insights into human biomechanics. Yet, currently available biomechanics datasets focus on either highly consistent, continuous, and symmetric activities, such as walking and running, or only a single specific non-cyclic task. To capture a more holistic picture of lower limb movements in everyday life, we collected data from 12 participants performing 20 non-cyclic activities (e.g. sit-to-stand, jumping, squatting, lunging, cutting) as well as 11 cyclic activities (e.g. walking, running) while kinematics (motion capture and IMUs), kinetics (force plates), and electromyography (EMG) were collected. This dataset provides normative biomechanics for a highly diverse range of activities and common tasks from a consistent set of participants and sensors. 

   more » « less
5. Design and Control of an Underactuated Exoskeleton for Loaded Walking

   Mau, Musharrat Mustaree; Asbeck, Alan T (April 2026, Journal of medical robotics research)

   This work presents a first demonstration of a nonanthropomorphic, underactuated exoskeleton architecture that uses a single linear actuator per leg to transfer loads directly to the ground next to the wearer’s foot. This allows the exoskeleton and user to move somewhat independently, facilitating normal kinematics during walking. We also present the control system that employs inertial measurement units (IMUs) affixed to the exoskeleton and shoes to estimate gait parameters. The performance of the exoskeleton is assessed based on its ability to touch down to the ground and lift off from the ground at the start of the stance and swing phases, respectively, as a participant walks on a treadmill at five speeds (0.67–1.34 m/s). The exoskeleton exhibited a touchdown delay of 0.093±0.019 seconds and a lift-off delay of 0.007±0.018 seconds, resulting in total support percentages of 88.9±2.9% during the stance phase across all speeds and the best performance at 0.67 m/s. Other experiments showed that our gait percentage estimation during steady state walking achieved a pooled root-mean-square error (RMSE) of 2.94%, which is less than prior works. This proof-of-concept architecture highlights a promising direction for lightweight, low-complexity exoskeletons for load carriage or bodyweight support. 

   more » « less