Taking measurements from two-dimensional image of a hand

Doctors looking after patients with hand injury often need to measure the recovery progress by measuring the angles at the finger joints. This is traditionally done using a goniometer to compare the range of motion of a joint at different stages of the treatment. However, this method requires the me...

Full description

Saved in:
Bibliographic Details
Main Author: Ng, Tsu Kian
Other Authors: Lee Yong Tsui
Format: Final Year Project
Language:English
Published: 2017
Subjects:
Online Access:http://hdl.handle.net/10356/72231
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:Doctors looking after patients with hand injury often need to measure the recovery progress by measuring the angles at the finger joints. This is traditionally done using a goniometer to compare the range of motion of a joint at different stages of the treatment. However, this method requires the medical practitioner to estimate the axis of rotation of a joint which often leads to inaccurate readings. Hence, this project aims to simplify and improve the consistency of such measurements using a computer-based approach to compute the angular measurements from the photograph of the hand. A user-controlled virtual hand model was created to allow the program user to replicate the patient’s actual hand position from a photograph. In doing so, the angles of the virtual hand will be similar to the actual hand. As the virtual hand was constructed using mathematical geometry, the angular measurements will be made known and displayed to the user. The results have shown the program’s ability to take consistent measurements of the hand with an absolute error of five degrees. The ability to modify the length of the virtual hand, increases the program’s usability to accommodate different hand sizes. However, more research regarding the variation in the hand anatomy across the globe needs to be carried out for future improvements of the project. The virtual hand controls and joint identification can be automated to minimize the need for human intervention and improve the reliability of the program.