Feasibility study of AI supported tool for respiratory difficulty
Study Period: January 2025-June 2026
Donor Name: University of Edinburgh, UK
Partners: Usher Institute, University of Edinburgh, UK
Project Description:
Background: Acute lower respiratory infections (ALRIs) remain a leading cause of childhood mortality and morbidity. Respiratory syncytial virus (RSV) is a major cause of ALRI, especially in children under 5 years of age. Early detection and triage of severe RSV enable timely referral and treatment, helping prevent avoidable hospitalization and death in children. However, this remains challenging due to limited access to trained health workers and diagnostic tools at the community level. Besides, there is no clear, validated method to measure disease severity in infants with RSV, which complicates clinical care. Artificial intelligence (AI) offers a new way to support community-level assessment in this context. An AI-supported tool using video recordings and a digital stethoscope could help detect disease severity in RSV, as well as other ALRIs in children.
Objective: The objective of the study is to assess the feasibility and acceptability of using an AI-supported tool to improve the identification of respiratory difficulty via video image recording and a digital stethoscope at community.
Method: We are conducting a qualitative study in Zakiganj Upazila (sub-district), Sylhet. It includes focus group discussions (FGDs) with service providers, caregivers, and community leaders, and in-depth interviews (IDIs) with health administrators and policymakers. We explain the AI-driven tool, including the digital stethoscope and video recording process, verbally before the interviews. The FGDs explore perceptions and acceptability of the tool, while the IDIs provide policy-level insights on its feasibility and implementation. All discussions take place in Bangla or Sylheti and are guided by a trained qualitative researcher and research assistant using a semi-structured interview guide.
Significance: The findings from this study will help meet the urgent need for reliable and accessible diagnostic tools for RSV and other ALRIs in children. Leveraging AI-driven insights, it will enhance early detection and management of respiratory distress, ultimately mitigating the impact of RSV and ALRIs on vulnerable pediatric populations.