A Multicenter Pilot Study of Biliary Atresia Screening Using Digital Stool Color Imaging

Area of Science:

• Medical Imaging
• Pediatric Gastroenterology
• Biomedical Engineering

Background:

• Biliary atresia (BA) diagnosis relies on screening tools like the stool color card (SCC).
• Early detection of BA is crucial for timely intervention, such as Kasai portoenterostomy.
• Current screening methods may lack objectivity and optimal sensitivity.

Purpose of the Study:

• To evaluate the efficacy of digital color image processing of infant stool images captured via smartphones for BA screening.
• To determine if digital color parameters offer a more objective and sensitive method for identifying BA stools.
• To compare the diagnostic accuracy of digital analysis with the traditional SCC.

Main Methods:

• A prospective study involving infant stool samples from five Thai hospitals.
• Digital image color analysis of stool photographs taken with smartphones.
• Development of software based on the SCC color scale for BA stool detection.
• Comparison of sensitivity and specificity between digital analysis and SCC.

Main Results:

• Digital color parameters, specifically saturation and blue values, showed potential in differentiating BA stools.
• Optimal cutoff thresholds (saturation ≤56 or blue ≥61) achieved high diagnostic accuracy (81.8% and 78.8%, respectively).
• Digital image processing demonstrated acceptable accuracy in identifying BA stools in jaundiced infants.

Conclusions:

• Digital image processing of smartphone-captured stool images is a promising technology for BA screening.
• Objective measurement of digital color parameters can improve the accuracy of BA stool identification.
• This method can help differentiate BA stools from those of non-BA infants with jaundice.