Performance of an artificial intelligence algorithm for interpreting lung sounds from children hospitalised with pneumonia in Malawi.
Nadia E Hoekstra, Maganizo B Chagomerana, Zachary H Smith...
https://pubmed.ncbi.nlm.nih.gov/40968644/Actively Recruiting
Led by Johns Hopkins University · Updated on 2026-06-08
2500
Participants Needed
1
Research Sites
N/A
Total Duration
Researchers are evaluating the use of an artificial intelligence AI-enabled digital stethoscope to improve antibiotic use in children aged 2 to 59 months with lower respiratory infections in rural Bangladesh. The study aims to determine if clinical guidelines enhanced by this digital stethoscope can safely reduce unnecessary antibiotic use while maintaining treatment effectiveness. The trial also assesses whether this approach is a cost-effective alternative to standard care, addressing the global problem of rising antibiotic resistance. Participants will be randomly assigned to one of two groups. One group will receive care guided by clinical guidelines enhanced by the AI-enabled digital stethoscope, which classifies lung sounds to decide if amoxicillin syrup or placebo should be given twice daily for five days. The other group will receive standard care with amoxicillin as per guidelines. Digital lung sound recordings are taken from four chest positions using a wireless stethoscope, and analysis is done by an automated algorithm. Masking is maintained by applying the stethoscope to all children and hiding the results. Children will be followed for two weeks with at least six study visits, including home visits by community health workers on days 2, 3, 5, and 14, and a clinic visit on day 7 to assess treatment success or failure. Caregivers can contact the study team anytime via a hotline. Safety is closely monitored, with referrals to hospital for treatment failure signs or low oxygen levels. The trial will track treatment failure, relapse, and safety outcomes, aiming to provide evidence on whether AI-enhanced diagnosis can safely reduce antibiotic use in children with pneumonia symptoms.
CONDITIONS
Trial of an AI-enabled Digital Stethoscope to Improve Antibiotic Stewardship
E
Eric D McCollum, MD, MPH
S
Salahuddin Ahmed, PhD
Explore thousands of other clinical trials that might be a better match.
Sign up to get personalized trial recommendations delivered to your inbox.
Already have an account? Log in here
Johns Hopkins University
Lead Sponsor
Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD)
Collaborating Sponsor
You may qualify if you...
You will not qualify if you...
Duration - 2 to 4 weeks
Participants are screened for eligibility to participate in the trial.
1 screening and enrollment visit (in-person)
Duration - 1 day
Participants will have lung sound recordings taken using a digital stethoscope to assist in antibiotic treatment decisions according to an automated algorithm.
1 visit (in-person)
Duration - 5 days
Participants receive either amoxicillin or placebo syrup twice daily for five days according to guidelines and randomized group assignment.
Daily administration over 5 days
Duration - 14 days
Participants are monitored through home visits by community healthcare workers and a study physician visit to assess treatment outcomes and safety, including assessment for treatment failure and adverse events.
Home visits on Days 2, 3, 5, and 14; study physician visit on Day 7
Study Type
INTERVENTIONAL
Masking
QUADRUPLE
Allocation
RANDOMIZED
Model
PARALLEL
Primary Purpose
DIAGNOSTIC
Number of Arms
2
Have more questions? Get in touch with our team for quick support
Nadia E Hoekstra, Maganizo B Chagomerana, Zachary H Smith...
https://pubmed.ncbi.nlm.nih.gov/40968644/Amy Sarah Ginsburg, Tisungane Mvalo, Evangelyn Nkwopara...
https://pubmed.ncbi.nlm.nih.gov/30419120/Eric D McCollum, Daniel E Park, Nora L Watson...
https://pubmed.ncbi.nlm.nih.gov/28883927/Salahuddin Ahmed, Saima Sultana, Ahad M Khan...
https://pubmed.ncbi.nlm.nih.gov/35493777/Dimitra Emmanouilidou, Eric D McCollum, Daniel E Park...
https://pubmed.ncbi.nlm.nih.gov/28641244/Tomasz Grzywalski, Mateusz Piecuch, Marcin Szajek...
https://pubmed.ncbi.nlm.nih.gov/30927097/Stephanie Graceffo, Amyna Husain, Salahuddin Ahmed...
https://pubmed.ncbi.nlm.nih.gov/33018152/Annapurna Kala, Amyna Husain, Eric D McCollum...
https://pubmed.ncbi.nlm.nih.gov/33018100/Dimitra Emmanouilidou, Eric D McCollum, Daniel E Park...
https://pubmed.ncbi.nlm.nih.gov/25879837/Daniel E Park, Nora L Watson, Christopher Focht...
https://pubmed.ncbi.nlm.nih.gov/35577452/Johns Hopkins University
Lead Sponsor
Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD)
Collaborating Sponsor