TB Chest X-Ray Diagnostic Technique Using Deep Learning

The bacteria Mycobacterium tuberculosis is responsible for the infectious illness tuberculosis (TB). Although the lungs are the primary organs affected, the kidneys, bones, and brain may also be affected. Whenever an infected person coughs, sneezes, or talks, small droplets comprising the bacteria are released into the air and spread through the airborne system. The elimination of tuberculosis (TB) remains a problem due to factors such medication resistance, co-infection with HIV, and scarce resources in high-burden configurations, even with substantial efforts to treat the disease. As a major worldwide health concern, tuberculosis (TB) necessitates quick and precise diagnosis in order to be effectively managed. This study aimed to develop a TB detection model using chest X-ray images obtained from Kaggle.com, utilizing Google's Collaboration Platform. The model was trained to recognize patterns within the TB chest X-rays to efficiently recognize TB within patients in order to be treated in time. We introduce a technique for transfer learning that proves preferable to the traditional methodology of adopting ImageN et weights. The hybrid methodology utilizes best first search technique in fusion with ResNet 50 model. The SVM classifier is used for categorizing the classes. An accuracy of 93.91 % is attained.