Computer-aided Detection of Tuberculosis from Microbiological and Radiographic Images

Abstract Tuberculosis caused by Mycobacterium tuberculosis have been a major challenge for medical and healthcare sectors in many underdeveloped countries with limited diagnosis tools. can be detected from microscopic slides chest X-ray but as result of the high cases tuberculosis, this method tedious both Microbiologists Radiologists lead to miss-diagnosis. These challenges solved employing Computer-Aided Detection (CAD)via AI-driven models which learn features based on convolution an output accuracy. In paper, we described automated discrimination microscope slide images into non-tuberculosis using pretrained AlexNet Models. The study employed Chest dataset made available Kaggle repository Near East University Hospital repository. For classification images, model achieved 90.56% accuracy, 97.78% sensitivity 83.33% specificity 70: 30 splits. 93.89% 96.67% 91.11% 70:30 Our is line notion that CNN used classifying higher accuracy precision.