Automated and non-contact human finger range of motion measurement system
No Thumbnail Available
Authors
Meeting name
Sponsors
Date
Journal Title
Format
Thesis
Subject
Abstract
[ACCESS RESTRICTED TO THE UNIVERSITY OF MISSOURI-COLUMBIA AT AUTHOR'S REQUEST.] Measurement of finger active range of motion (ROM) is essential for monitoring the effectiveness of rehabilitative treatments and for evaluating patients' functional impairment accurately. Currently, finger ROM is measured through a labor-intensive process of applying a hand-held goniometer to each finger joint and recording the value. This method is subject to error and is time-consuming, leading many surgeons not to collect the data to avoid delays in the clinic. To speed up and simplify this process, we proposed a system to measure the ROM of joints of each finger automatically. The system is based on extracting a 3D hand model using the low-cost Intel RealSense SR300 camera that can produce an accurate 3D model. Segmentation methods are developed to extract each finger individually, and several algorithms are proposed to estimate joint angles. To evaluate the proposed system, we collected data on 30 healthy volunteers and 22 hand therapy patients, with UNIVERSITY OF MISSOURI-COLUMBIA IRB approval. The system has been tested and compared to manual measurements made with a goniometer by a fellowship trained hand surgeon to extract the finger joint ROM. First, using data on the healthy subjects, the mean absolute differences in measurement were 8 degrees across all finger joints. These differences were found to compare favorably to the variability noted in manual measurements in published papers. Moreover, the proposed system estimates MCP and PIP joint angles in all long fingers accurately with one automated, non-contact scan. The system was improved and then further tested with patients in the hand therapy clinic, who may have finger swelling, attachments, and/or unusual finger ROM. The system showed a mean absolute difference of 7 degrees with respect to the goniometer readings. Moreover, analyzing the clinic use provided recommendations for further development of the proposed system.
Table of Contents
DOI
PubMed ID
Degree
Ph. D.
Thesis Department
Rights
Access to files is limited to the University of Missouri--Columbia.