Published Paper Details:

Diabetes mellitus is a metabolic condition characterized by persistently elevated blood glucose levels (DM). Hyperglycemia, the medical term for abnormally high blood glucose levels, is a hallmark of diabetes mellitus. Diabetes mellitus (DM) develops from a combination of hereditary and environmental factors that lead to immune system dysfunction and insulin resistance. Factors that increase the likelihood that you may get diabetes include having a high body mass index or having a first-degree relative who has the disease (BMI) The global prevalence of obesity has risen as a result of poor dietary practices and an increase in sugar intake. Statistics all throughout the globe are showing the effects. The global prevalence of diabetes is increasing, with an estimated 1.6 million people having the disease by the year 2025. Diabetic microvascular problems, such as diabetic foot, may be prevented with early diagnosis and appropriate treatment, hence voluntary and automated procedures for detecting DM are crucial. Alternatively, foot thermography is a potential technology because it enables viewing of thermal patterns, patterns that are changed due to shear and friction associated with lower limb neuropathy. For the purpose of synthesizing thermal images of the foot, we use the usage of categorization indices such as the Thermal Change Index (TCI). In light of these considerations, in this investigation we use deep convolutional neural networks, using 12 distinct data augmentation approaches, to identify DM patients based on foot thermography alone.