@ijtsrd

Segmentation and Classification for Hyperspectral Imaging of Foot Inspection in Vascular and Neuro Images

. INTERNATIONAL JOURNAL OF TREND IN SCIENTIFIC RESEARCH AND DEVELOPMENT, 7 (1): 1114-1122 (февраля 2023)

Аннотация

Diabetes Mellitus DM and Diabetic Neuropathy DN are the most common diseases in the worldwide, according to the World Health Organization WHO . A high index of death is also correlated with it. Diabetic neuropathy is a significant worldwide cause of neuropathy which can lead to amputations and disabilities. Diabetic neuropathy can have multiple clinical manifestations, the most common presentation being distal symmetric polyneuropathy and the key mechanism for diabetic foot development. One of the major problems is diabetic foot, which includes the creation of plantar foot hyper spectral which may result in amputation. Several studies report that hyperspectral is helpful in identifying differences in plantar temperature, which may lead to a higher risk of ulceration. However, in diabetic patients, the distribution of plantar temperature does not follow a standard sequence, thereby making it impossible to quantify the changes. There is also an importance in enhancing the performance of the methods of analysis and classification that help to diagnose abnormal variations in the temperature of the plantar. All this refers to the use of computer aided programmes that work with extremely structured data structures, such as those involved in artificial intelligence AI . This study combines approaches based on machine learning with Deep Learning DL structures. Furthermore, we developed a new DL structure, which is qualified and is able to achieve higher significance in terms of precision and other quality metrics. The key purpose of this study is to examine the use of AI and DL for the classification of hyperspectral images of the diabetic foot, demonstrating its advantages and disadvantages. To the best of our understanding, this is the first suggestion for the definition of diabetic foot hyperspectral implemented by DL networks. The studies are carried out in DM and control groups through hyperspectral images. Afterwards, based on a pre reported hyperspectral shift index, a multi level classification is done. The high precision attained illustrates the utility of AI and DL as auxiliary instruments to help in medical diagnosis. The aim of this study was to perform a systematic and updated analysis of diabetic neuropathy, concentrating on its classification, diagnostic research and treatment. Dr. C. Yesubai Rubhavathi | Diofrin. J | Eben Paul Richard. S | Arunachalam. R | Vishnu Durga. S "Segmentation and Classification for Hyperspectral Imaging of Foot Inspection in Vascular and Neuro Images" Published in International Journal of Trend in Scientific Research and Development (ijtsrd), ISSN: 2456-6470, Volume-7 | Issue-1 , February 2023, URL: https://www.ijtsrd.com/papers/ijtsrd52793.pdf Paper URL: https://www.ijtsrd.com/engineering/computer-engineering/52793/segmentation-and-classification-for-hyperspectral-imaging-of-foot-inspection-in-vascular-and-neuro-images/dr-c-yesubai-rubhavathi

Линки и ресурсы

тэги