Global Journal of Computer Science and Technology, C: Software & Data Engineering, Volume 22 Issue 2

relevant information to detect blood vessels. They then performed manual cropping and used adaptive histogram equalization to improve the image’s contrast. They then used image enhancement to segment the image using a Gaussian filter followed by OTSU thresholding. They then used PrincipalComponent Analysis(PCA) to extract the features. A Support Vector Machine then performed the classification. They reported accuracy of 83.3% but not the time needed for automated analysis. ending with results. They reported an overall accuracy of 83.3% but not the time needed for automated analysis. In their next paper [48], they experimented with different types of machine learning techniques, including Random Forests Classifier, Multinomial Logistic Regression, and CNNs. However, they do not report the timing needed for classifying the blood vessels. Perikumar Java et al. [49] used a custom form of ResNet18 [50] to quantify capillaries. They used a 10- layer architecture and resized the images to input 224x224x3. They (a) (b) (c) (d) Capillary X: A Software Design Pattern for Analyzing Medical Images in Real-time using Deep Learning Global Journal of Computer Science and Technology Volume XXII Issue II Version I 16 Year 2022 ( ) C © 2022 Global Journals KV Suma et al. [47] used Fuzzy Logic Kernels to classify the images. They started by Fuzzification of the input, followed by the Application of the Fuzzy operator, then aggregating the consequents across the rules,