Paper Info
Title
Early detection of foot ulceration in type II diabetic patient using registration method in infrared images and descriptive comparison with deep learning methods
Abstract
In today’s scenario, Diabetes Mellitus (DM) is a major medical issue that comes with numerous complications. It is the most imperative disease associated with a diabetic foot. Among other conditions, one of the most common symptoms of diabetes is a diabetic foot ulcer. The early diagnosis of such foot perplexities, which may later lead to foot removal, can shield diabetic patients from many critical stages. The main objective of this work relies on the area of early diagnosis and minimizing the occurrence of ulcers in the diabetic foot. Infrared imaging is an appropriate tool that extracts information for the determination of various diseases. The primary purpose of applying infrared imaging in a diabetic patient is for the early diagnosis of ulcers. This type of diagnosis is non-invasive and faster than colored imaging techniques. This work focuses on temperature variation in the feet of more than 60 persons (37 men (61.67%) and 23 women (38.33%)). The mean age of persons for detecting ulcer formation due to diabetes is 60.50 ± 16.71 years. The foot is divided into six major segments, followed by close observation of the individual part’s temperature variation. In the proposed work, image registration is applied to measure the temperature difference between the regions of feet, assuming the threshold temperature as 2.2 °C. The results are obtained using the image processing toolbox of MATLAB. After experimental operations, the results clearly distinguish the foot region, having a temperature difference higher than the assumed threshold value. This analysis distinctly classifies the ulcer risk foot and is quite easy to understand as compared with existing Deep Learning techniques. The proposed methodology is also very less complex in implementation as compared to those of Deep Learning methods. It provides a non-invasive technique for the diagnosis of foot ulcers in a diabetic patient using infrared imaging.,
Year
DOI
Venue
2022
10.1007/s11227-022-04380-z
The Journal of Supercomputing
Keywords
DocType
Volume
Diabetes mellitus, Non-invasive, Type II diabetes, Ulceration risk, Diabetic foot, Deep learning, Infrared image
Journal
78
Issue
ISSN
Citations 
11
0920-8542
0
PageRank 
References 
Authors
0.34
3
4
Name
Order
Citations
PageRank
Mritunjay Rai100.34
Tanmoy Maity200.34
Rohit Sharma300.34
R. K. Yadav400.68