In 2018, around 1.5 million people died from tuberculosis (TB) — an infectious disease that mainly affects the lungs. A major obstacle in the clinical treatment of TB is the long therapy time required to clear the infection. An infected patient needs to take antibiotics for over 6 to 9 months to prevent a relapse — a duration so long that many discontinue their medications.
IIT Bombay’s new deep learning framework, named SpADANet, enhances damage classification accuracy using limited labels across multiple hurricanes.