Investigator(s)
| WHO Technical Officer |
Katrin Bote WHO SEARO -Neglected Tropical Diseases Katrin Bote 
|
| Principal Investigator |
Priti Meshram The Grant Medical College & Sir J.J. Group of Hospitals, Mohammed Ali Rd, Noor Baug, Mazgaon, Mumbai, Maharashtra 400003 Priti Meshram
|
Title(s) and abstract
| Scientific title | Evaluating the Usefulness of AI-based Chest X-ray Screening in Enabling Incidental Diagnosis of Tuberculosis in a Hospital setting |
| Public title | Evaluating the Usefulness of AI-based Chest X-ray Screening in Enabling Incidental Diagnosis of Tuberculosis in a Hospital setting |
| Background | Tuberculosis (TB) is an infectious communicable disease which remains a major cause of morbidity and mortality worldwide. As per WHO estimate 2021, over 10.6 million individuals were infected with TB worldwide and out of which around 28% are from India. As per the India TB report 2023, a 13% increase in TB notifications (24.2 lakh cases) and highest TB case notification rate (172 cases per lakh population) was marked in the year 2022. Furthermore, the prevalence survey by India’s National Tuberculosis Elimination Program (NTEP) revealed the significance of using chest X-ray examinations to identify signs suggestive of TB and its potential in finding incidental cases to overcome any delay in TB diagnosis and treatment cascade. |
| Objectives | Though chest X-rays were widely used as a screening tool for pulmonary TB, in case of low resource settings, the rising population, scarcity of qualified radiologists, radiologists on rotation can lead to heavy radiographic workload, backlog and missed diagnoses. To tackle these shortcomings, the adoption and integration of Artificial Intelligence (AI) based deep learning analysis within the diagnostic workflow has been proven helpful. With increasingly promising evidence towards the usefulness of AI in screening and triage of TB, WHO has now recommended their use as an alternative to human interpretation of digital CXR for screening and triage for TB in adults aged 15 years or older. |
| Study Methods | However, targeted operational studies assessing the usefulness and impact of AI based TB screening and triaging tools are necessary in planning effective TB control and preventive strategies. In this study, we aim to assess and test the performance of qXR (AI based CXR processing software medical device) in identifying incidental cases with pulmonary tuberculosis in a public hospital setting. |
| Expected outcomes and use of results | AI can enable screening of all the chest X-rays in for signs of TB (in high TB burden settings) including Chest X-Rays from non TB pathways (such as pre-surgical Chest X-Rays), thus detecting incidental cases that are often missed owing to lack of symptoms or awareness. AI also has the potential to bring in larger numbers to the screening fold by establishing continuous passive screening of Chest X-Rays for TB prevalent settings. The goal of the study is to generate evidence about usefulness of an AI-based computer assisted detection device (qXR) in a government hospital for AI-based chest X-ray interpretation for identification of incidental patients with pulmonary tuberculosis. |
| Keywords | tuberculosis, diagnosis, screening, artificial intelligence |
Research Details
| Student research | No |
| Start Date | 01-Jan-2024 |
| End Date | 31-Dec-2024 |
| Key Implementing Institution | Grant Medical College & Sir J.J. Group of Hospitals, Mumbai |
| Multi-country research | No |
| Nationwide research | No |
| India | |
| Research Domain(s) | Communicable Disease Research |
| Research field(s) | Tuberculosis |
| Involves human subjects | Yes |
| Operational Research | |
| Data Collection | Primary data |
| Proposal reviewed by other Committee | No |