Current Pedorthics | January-February 2020 | Vol.52, Issue 1 | Page 30

Mobile Application for Ulcer Detection as listed in Fig. (4). The background thread protects the main thread from being clogged up with processing techniques, which helps in providing a smooth flowing user experience. 3. RESULTS The result shown to the user is the overall view of his feet with the help of a BGR image, as shown in Fig. (7). The blue color on the screen represents the cold segments on the image, and the red region represents the hot segment, which indicates the occurrence of DFU. The Mean Temperature Difference (MTD) is provided to the user, which is the difference between the suspected hottest region and the feet region. If the MTD is more than 2.2 o C, the application highlights the BGR image with red color on the suspected ulcerous region. Four test images were used to test the algorithm developed using OpenCV software. The first image is for a nondiabetic feet patient with no abnormalities. The other three images Figure 8: Test Image 1 28 Pedorthic Footcare Association | www.pedorthics.org are with simulated ulcer on different regions. The four test images are shown in Figs. (8, 9, 10, and 11) where the mean temperature values for the feet segments (hot minus hottest) are 33.2 o C, 36.9 o C, 35.5 o C, and 35.3 o C, respectively. The suspected hottest region mean temperature values are 35.0 o C, 39.4 o C, 38.0 o C, and 37.7 o C, respectively. If an ulcer occurs, the application automatically highlights the region with a red color over the BGR blue feet segment image. The MTD for the first test image shown in Fig. (8) is 1.8 o C, while the MTD values for the rest of images containing a simulated ulcer region are 2.5 o C, 2.5 o C, and 2.4 o C, respectively. Table 1 depicts the observed values for the four test images. 4. DISCUSSION The proposed mobile thermal imaging system represents a new promising approach to detect possible ulcers in the diabetic patients’ Figure 9: Test Image 2