Paediatric TB case finding and the role of AI

The conversation between Dr. Shibu Vijayan and Dr. Jagannath Devan at The Union 2023 focuses on pediatric tuberculosis (TB) diagnosis, the development of new diagnostic tools, and the role of artificial intelligence (AI) in improving pediatric TB case finding. 

Dr. Shibu Vijayan, Medical Director at Qure.ai, interviews Dr. Jagannath Devan, an Assistant Professor of Pediatric Infectious Diseases at the University of California, San Francisco. Dr. Devan is also a clinical researcher and data scientist with a focus on pediatric TB diagnosis. 

1. Innovations in Pediatric TB Diagnosis: Dr. Devan highlights ongoing research in developing more feasible testing methods for pediatric TB, including swab and stool-based tests. He discusses efforts to adapt existing adult-focused tests for children, such as urine-based tests using markers like LAMP. 

2. AI in Pediatric TB Diagnosis: The role of AI in interpreting chest X-rays for TB diagnosis in children is explored. Given the unique presentation of TB in children and challenges in obtaining quality X-ray images, Dr. Devan sees AI as a valuable tool for reliable diagnosis and triage. 

3. Data Repositories for AI Validation: The importance of building well-characterized image repositories for validating and fine-tuning AI algorithms for pediatric use is discussed. Dr. Devan mentions the Capture Consortium's efforts in this area. 

4. Sound-Based Diagnostic Tools: Dr. Devan also touches on the potential of cough and lung sound-based tools as non-invasive, simple methods for pediatric TB diagnosis, emphasizing the need for child-specific algorithms. 

5. Broader Applications of AI in Pediatric Care: The conversation extends to the broader application of AI in pediatric infectious diseases, stressing the need for solutions that address a range of respiratory conditions beyond TB, considering the nonspecific symptoms with which children present. 

The discussion concludes with Dr. Devan emphasizing the potential of AI in pediatric infectious diseases, not just for TB but for a wider range of respiratory conditions. He advocates for developing AI tools that aid frontline providers in making critical decisions in pediatric care, highlighting the importance of interdisciplinary collaboration and continued research in this field.