A novel mobile phone application for infant stool color recognition: An easy and effective tool to identify acholic stools in newborns

Abstract

Objectives Early diagnosis of biliary atresia is essential to improve long-term outcomes. Newborn screening with an infant stool color card allows early recognition of biliary atresia patients. Our aim was to develop and validate a mobile phone application (PopòApp) able to identify acholic stools. Methods An intuitive app was developed for iOS and Android smartphones. A learning machine process was used to generate an algorithm for stools color recognition based on the seven colors of the infant stool color card, which were considered as the gold standard. Consecutive images of stools were taken by the PopòApp, directly into the diapers of children aged ≤6 months. The PopòApp classified the photographs as “normal”, “acholic” or “uncertain”. To validate the PopòApp, four doctors independently classified all images, and only those for which all doctors agreed were included. The sensitivity, specificity, positive/negative predictive values, and accuracy of the PopòApp were evaluated. Results Of 165 images collected, 160 were included in the study. All acholic stools were recognized by the PopòApp. The PopòApp sensitivity was 100% (95% CI:93.9%–100%) with no false negatives, regardless of the brand of phone. The specificity was 99.0% (95% CI:94.6%–99.9%). The accurancy of the PopòApp was 99.4% (95% CI:96.6%–99.9%), with a positive predictive value of 98.4% (95% CI:89.8%–99.8%). Conclusion The current study proved, in a large cohort, that the PopòApp is an accurate and easy tool for recognition of acholic stools. The mobile App may represent an effective strategy for the early referral of children with acholic stools, and potentially could improve the outcomes of biliary atresia.

Machine Learning Evaluation of Biliary Atresia Patients to Predict Long-Term Outcome after the Kasai Procedure

“…In detail, Hoshino et al, using an ML algorithm, realized an iPhone application (Baby-Poop) able to capture subtle differences in stool color that may be undetectable by a layperson to get early diagnosis of BA [ 55 ]. A similar ML application with the same purpose was made by Angelico et al, who created PopòApp [ 56 ]. Moreover, Zhou et al developed an ensembled deep learning model to facilitate the diagnosis of BA for non-expert radiologists using DB values and US images as well as videos of the gallbladder [ 57 ].…”

Biliary Atresia in 2021: Epidemiology, Screening and Public Policy

“…Popòapp ® , designed by an Italian team, is another mobile device application for both iOS and Android devices [43]. The colour analyzer algorithm is based on the Japanese seven-stool colour photo panel using an RGB digital colour system.…”

“…Mobile device screening application (Angelico et al[43], © 2017 reprinted by the permission of Sage Publications).…”

Neonatal Jaundice Detection Using a Computer Vision System

“…Angelico et al [29] developed an intuitive application for Android smartphones. Machine learning was used in the application to generate an algorithm for stool color recognition based on the seven grades of the infant stool color card, which was considered as the reference model.…”