TESTING AND VALIDATION OF MACHINE LEARNING MODEL FOR TUBERCULOSIS SCREENING USING CHEST RADIOGRAPHS IN MALAYSIAN POPULATION
NMRR ID-23-00114-ETO
iTB-REPOXR, part of the broader iTB project, is a pioneering initiative that leverages advanced technology to transform medical diagnosis, while also contributing to the global and national efforts to end tuberculosis by 2035. This endeavor focuses on creating a comprehensive repository consisting of chest radiographs and associated clinical information. The primary purpose of this repository is to facilitate the development of machine learning techniques specifically tailored for medical image analysis, with a primary emphasis on diagnosing lung diseases, including the prominent case of tuberculosis.
The core essence of iTB-REPOXR lies in its foundational role, laying the groundwork for a more ambitious undertaking known as iTBXR. This subsequent phase is characterized by the aspiration to construct an artificial intelligence (AI) model of remarkable sophistication, engineered to scrutinize and diagnose an extensive array of lung-related ailments. The significance of this endeavor is underscored by its objective to identify not just tuberculosis but a diverse spectrum of at least 12 distinct lung diseases and disorders. By creating a vast reservoir of chest radiographs supplemented with clinical insights, this initiative provides fertile ground for the evolution of AI-powered medical image analysis. Such advancements have the potential to redefine the landscape of healthcare, offering enhanced accuracy and efficiency in disease diagnosis and treatment planning.
By establishing a comprehensive repository with iTB-REPOXR and validating machine learning algorithms for lung disease diagnosis, it paves the way for iTBXR's more ambitious aim of creating an AI model capable of identifying a diverse array of lung-related conditions. Through these initiatives, the healthcare industry stands poised to embrace a new era of precision medicine and improved patient care.