Tuberculosis (TB) is the leading cause of death from a single infectious disease in the world, above HIV/AIDS. This disease is transmitted through the inhalation of particles from coming in close contacts with infected individuals. In 2019 alone, 1.4 million lives were lost due to TB. This number is concerning since the disease is curable and preventable.
As part of the Sustainable Development Goals, the United Nations and World Health Organization have pledged to end the TB epidemic by 2030. However, the breakout of the COVID19 pandemic in 2019 has resulted in human, financial, and other resource reallocation that reversed the effects of previous efforts. Recently for the first time, WHO has recommended the use of computer-aided assistant to help fill in the gap in resources to accelerate TB screening and triage.
Locally, in the UAE, TB screening has been made mandatory for expatriates before they can establish residency in the country. Misdiagnoses are undeniably expensive since false negatives pose an imminent danger to the population, while false positives translate into lost opportunities and potential talents in the UAE. This project aims to create a reliable and trustworthy computer-assisted diagnostic application to assist radiologists in the detection and localization of TB from chest X-rays.