The Dirty Nappy Study
Research type
Research Study
Full title
The Dirty Nappy Study: Development and Evaluation of a Machine Learning Algorithm for the Early Detection of Cholestasis and Biliary Atresia from Parent-Provided Images of Dirty Nappies
IRAS ID
327150
Contact name
Girish Gupte
Contact email
Sponsor organisation
Birmingham Women’s and Children’s NHS Foundation Trust
Duration of Study in the UK
1 years, 6 months, 0 days
Research summary
Cholestasis in infants is a condition characterised by the impaired flow of bile from the liver to the intestines, leading to its accumulation in the liver, which can cause liver damage and other health issues. This condition, often manifesting shortly after birth with symptoms such as persistent jaundice & pale stools, can arise from various causes, including biliary atresia (BA), the most critical cause. BA needs surgery by 30 days of life to avoid severe liver damage and potential liver transplantation, hence an early diagnosis is important.
This novel study aims to harness machine learning for the early detection of cholestasis and biliary atresia, using images of infants' dirty nappies provided by parents. Conducted across four UK hospitals, the research will involve collecting smartphone images of 5 dirty nappies from parents of newborns at 14, 21, and 28 days of life, then at 3 and 6 months. These images will be analysed using machine learning to classify them as normal or abnormal, identifying infants at risk for biliary atresia.
This machine-learning algorithm's accuracy in detecting cholestasis and biliary atresia through such images will be evaluated, alongside assessing the feasibility and acceptability of this approach as a screening tool. The study seeks to gather a comprehensive dataset from both cholestatic and healthy infants, aiming for a robust evaluation of the algorithm's sensitivity, specificity, and overall accuracy. Expected to collect around 59,200 images from both affected and healthy infants, the research anticipates offering a significant advancement in early detection methods.
By facilitating early intervention, this approach has the potential to markedly improve outcomes for children with liver diseases. The findings, aimed for dissemination through peer-reviewed journals and conferences, could pave the way for a new screening tool for cholestasis and biliary atresia, significantly impacting paediatric liver disease management.
REC name
West Midlands - Coventry & Warwickshire Research Ethics Committee
REC reference
24/WM/0176
Date of REC Opinion
7 Oct 2024
REC opinion
Further Information Favourable Opinion