Researchers Manaswini Davuluri and Venkata Sai Teja Yarlagadda Leverage AI to Revolutionize TB Screening
North Carolina, USA – In a major advancement for tuberculosis (TB) diagnosis, researchers Manaswini Davuluri, Sr. Project Manager and Research Scientist at the Department of Information Systems in North Carolina, and Venkata Sai Teja Yarlagadda, Sr. Researcher and Software Engineer at Indiana Wesleyan University’s Department of IT, have developed an AI-powered device aimed at faster, more accurate TB screening. This novel technology utilizes convolutional neural networks (CNNs) to detect TB-related abnormalities in chest X-rays, offering a potentially transformative tool for healthcare systems globally.
TB, an infectious disease caused by Mycobacterium tuberculosis, remains a leading cause of illness and death worldwide, especially in low- and middle-income countries. The World Health Organization (WHO) reports that in 2021, 10.6 million people contracted TB, with 1.6 million succumbing to the disease. Factors such as delayed diagnosis, lack of healthcare resources, and drug-resistant TB strains have contributed to the persistence of this health crisis.
The new device addresses these challenges by employing deep learning to enhance TB detection. The research team focused on CNN algorithms to create an automated system for identifying TB symptoms from chest radiographs, providing a more efficient alternative to traditional diagnostic methods. Their study compared this model's accuracy, sensitivity, and specificity against conventional TB diagnostic approaches, showing marked improvements in both speed and precision.
“We recognized a gap in TB diagnosis—traditional methods can be slow and often inaccessible to populations most in need,” said Manaswini Davuluri, whose eight years of experience in AI project management and data-centric research has fueled her commitment to advancing healthcare solutions. “By applying AI to automate workflow, streamline decision-making, and optimize data processes, we aim to reduce delays in diagnosis and bring TB screening to more people.”
The integration of deep learning into tuberculosis (TB) screening marks a major leap forward, offering advancements that could reshape diagnostic practices. One of the key benefits of this new AI-driven device is its ability to deliver rapid diagnoses. By processing results in seconds, it enables faster intervention in TB cases, which is especially crucial in high-burden regions where timely treatment can help prevent the spread of the disease and improve patient outcomes.
Another significant advantage of this technology is its accessibility for low-resource settings. The AI model is compatible with mobile and portable imaging devices, making it viable for TB screening in remote or underserved areas where access to sophisticated diagnostic equipment is often limited. This accessibility can bridge healthcare gaps and bring vital screening capabilities to areas in need.
In addition to reaching more patients, the AI device provides essential support for healthcare professionals. Serving as a diagnostic aid, it assists radiologists and clinicians by flagging potential TB cases, thus reducing the chances of human error. This allows healthcare professionals to focus their time and resources on confirmed cases, enhancing workflow efficiency and patient care.
The AI device is also cost-effective, as it automates the screening process and minimizes the need for labor-intensive traditional methods. This automation could lead to cost savings for healthcare systems, making TB screening more affordable and enabling better resource allocation.
As AI continues to make strides in medical imaging, its applications are expanding beyond TB to other diseases, such as lung cancer and pneumonia, demonstrating its broader potential in healthcare. While the technology holds substantial promise, there are still challenges to address. Issues like data bias, model interpretability, and clinical validation are critical for ensuring the device’s effective integration into healthcare systems. Addressing these challenges will be essential to fully harness the potential of AI in revolutionizing disease diagnosis and improving public health outcomes on a global scale.
“This project underscores the potential of AI in making healthcare more accessible and efficient,” said Yarlagadda, a young researcher from Indiana Wesleyan University. “By innovating in TB diagnostics, we hope to contribute meaningfully to the global fight against this disease.”
As TB continues to pose a global health challenge, this AI-driven device represents a promising step forward. The research of Davuluri and Yarlagadda has opened up new possibilities for faster, accurate diagnoses and has the potential to reshape TB control efforts worldwide, ultimately improving health outcomes for millions affected by TB.
Contact Information:
Manaswini Davuluri
Sr. Project Manager and Research Scientist
Department of Information Systems, North Carolina, USA
Email: manaswini.m17@gmail.com
Venkata Sai Teja Yarlagadda
Sr. Researcher and Software Engineer
Department of Information Technology, Indiana Wesleyan University, Marion, Indiana, USA
Email: yarlagadda.teja@myemail.indwes.edu