Identification of Diabetic Foot Ulcer by Using Filter and Pixel Classification Methods

Abstract

Ulceration of the foot in diabetes is common and disabling and frequently leads to amputation of the leg. Mortality is high and healed ulcers often recur. The pathogenesis of foot ulceration is complex, clinical presentation variable, and management requires early expert assessment. Interventions should be directed at infection, peripheral ischaemia, and abnormal pressure-loading caused by peripheral neuropathy and limited joint mobility. Despite treatment, ulcers readily become chronic wounds. Diabetic foot ulcers are sores that occur on the feet of people with type 1 diabetes and type 2 diabetes. Up to 25% of people with diabetes will be affected by foot problems. Diabetic foot ulcers usually occur on the bottom of the foot. The main objective of this project is to identify the risk of diabetic foot and also classify the diabetic foot into  possible classes: “ulcus,” “callus,” “fissure” and “healthy.” By applying different digital image processing techniques like Sobel filter and pixel classification methods were used to identify the diabetic foot ulcer and classify the foot ulcer into “ulcus,” “callus,” “fissure” and “healthy” portions. First, we apply the filter on a diabetic foot image to detect its edges. Now, for that image we identify the pixels, and also compare each pixel with its next pixel. Based on the intensities of those pixels a new image with different colors is formed. Based on those colors is classified the diabetic foot into above mentioned categories. If the color is black that is “ulcus,” if the color is yellow that is “fissure,” if the color is red that is “callus,” and finally if the color is blue the portion is healthy one. Diabetic foot ulcers have been neglected in healthcare research and planning, and clinical practice is based more on opinion than scientific fact. Furthermore, the pathological processes are poorly understood and poorly taught and communication between the many specialties involved is disjointed and insensitive to the needs of patients. Infection of diabetic foot ulcers has particular risk of involvement of the adjacent bone resulting in serious complication of osteomyelitis. This project is to identify the osteomyelitis by long-term study of above method.

Keywords: Diabetic, digital image processing (DIP), FPG (fasting plasma glucose), impaired fasting glucose (IFG), OGTT (oral glucose tolerance test)

Cite this Article:

Srinivasulu Munakala, Sumathi P. Identification of diabetic foot ulcer by using filter and pixel classification methods. Research & Reviews: A Journal of Bioinformatics. 2015; 2(1): 22–31p.