DEVELOPMENT AND CLINICAL VALIDATION OF THE SENSOR-E EXOSKELETON FOR OBJECTIVE HAND FUNCTION ASSESSMENT IN STROKE REHABILITATION

Author

Madison Bates, University of KentuckyFollow

Author ORCID Identifier

https://orcid.org/0000-0003-1009-0006

Date Available

6-5-2026

Year of Publication

2025

Document Type

Doctoral Dissertation

Degree Name

Doctor of Philosophy (PhD)

College

Engineering

Department/School/Program

Biomedical Engineering

Faculty

Sridhar Sunderam

Abstract

Hand function is often impaired following neurological trauma or in conditions such as stroke, limiting independence and the ability to perform activities of daily living (ADLs). Current clinical assessments rely heavily on subjective rating scales that lack the resolution to detect subtle changes in motor performance, which can limit access to rehabilitation. Recent advances in wearable technologies highlight the potential for hand-worn devices to provide objective, real-time measures of hand function. However, most systems focus solely on kinematics and rarely capture both movement and exerted force. Moreover, integration into rehabilitation frameworks and brain-computer interface (BCI) protocols remains limited. To address these gaps, we have developed the Sensor-E Exoskeleton (SEE), a lightweight, wearable device that integrates flexion and force sensors to simultaneously monitor finger flexion and fingertip force. In a validation study with thirty non-clinical, healthy participants, SEE reliably distinguished graded levels of finger extension, contraction, and force exertion. Outputs were strongly correlated with reference measurement systems, including motion capture and load cell, with mean relative errors below 1%. Participants also interacted with a graphical user interface (GUI) that provided real-time feedback, enhancing accuracy and engagement. Building on these results, SEE was evaluated in stroke survivors (n = 3) and age- and hand dominance-matched controls (n = 18) as they completed both graded hand tasks and four functional ADLs (e.g., grasping, drinking, twisting, and pulling/pushing). Analyses revealed the device’s ability to capture differences between impaired and unimpaired hands and detect coordination strategies. Machine learning models achieved high accuracy with substantial agreement for task recognition and handedness (Cohen’s Kappa (Κ) 0.81-0.87) and moderate agreement for impairment detection across task (Κ 0.50-0.70). Although sensitivity in impairment classification was limited by the small stroke sample, findings highlight SEE’s potential to detect specific impairments. Collectively, this work establishes SEE as a reliable and versatile platform for objective assessment of hand function. SEE demonstrates clear potential as both a clinical assessment tool and a rehabilitation aid by simultaneously capturing movement and force during controlled and functional tasks.

Digital Object Identifier (DOI)

https://doi.org/10.13023/etd.2025.549

Funding Information

This work was supported by National Science Foundation Grant (no.: 1849213) from 2020 to 2023, the Halcomb Fellowship in Medicine and Engineering at the University of Kentucky (to Madison Bates) from 2023 to 2025, and NIH National Center for Advancing Translational Sciences (no.: UL1TR001998) in 2025.

Recommended Citation

Bates, Madison, "DEVELOPMENT AND CLINICAL VALIDATION OF THE SENSOR-E EXOSKELETON FOR OBJECTIVE HAND FUNCTION ASSESSMENT IN STROKE REHABILITATION" (2025). Theses and Dissertations--Biomedical Engineering. 88.
https://uknowledge.uky.edu/cbme_etds/88