Am J Gastroenterol. 2022 Mar 14. doi: 10.14309/ajg.0000000000001723. Online ahead of print.

Mark Pimentel ^1 2, Ruchi Mathur ^1 3, Jiajing Wang ¹, Christine Chang ¹, Ava Hosseini ¹, Alyson Fiorentino ¹, Mohamad Rashid ¹, Nipaporn Pichetshote ², Benjamin Basseri ², Leo Treyzon ², Bianca Chang ², Gabriela Leite ¹, Walter Morales ¹, Stacy Weitsman ¹, Asaf Kraus ⁴, Ali Rezaie ^1 2

Author information

¹Medically Associated Science and Technology (MAST) Program, Cedars-Sinai, Los Angeles, California, USA.

²Karsh Division of Gastroenterology and Hepatology, Cedars-Sinai, Los Angeles, California, USA.

³Division of Endocrinology, Diabetes and Metabolism, Cedars-Sinai, Los Angeles, California, USA.

⁴Dieta Health, Oak Park, California, USA.

Abstract

Introduction: Stool form assessment relies on subjective patient reports using the Bristol Stool Scale (BSS). In a novel smartphone application (app), trained artificial intelligence (AI) characterizes digital images of users' stool. Here we evaluate this AI for accuracy in assessing stool characteristics.

Methods: Subjects with diarrhea-predominant irritable bowel syndrome image-captured every stool for 2 weeks using the app, which assessed images for 5 visual characteristics (BSS, consistency, fragmentation, edge fuzziness, and volume). In the validation phase, using 2 expert gastroenterologists as gold standard, sensitivity, specificity, accuracy and diagnostic odds ratios of subject-reported vs AI-graded BSS scores were compared. In the implementation phase, agreements between AI-graded and subject-reported daily average BSS scores were determined, and subject BSS and AI stool characteristics scores were correlated with diarrhea-predominant irritable bowel syndrome symptom severity scores.

Results: In validation-phase (n = 14), there was good agreement between 2 experts and AI characterizations for BSS (intraclass correlation coefficients [ICC] = 0.782-0.852), stool consistency (ICC = 0.873-0.890), edge fuzziness (ICC = 0.836-0.839), fragmentation (ICC = 0.837-0.863), and volume (ICC = 0.725-0.851). AI outperformed subjects' self-reports in categorizing daily average BSS scores as constipation, normal, or diarrhea. In implementation-phase (n = 25), agreement between AI and self-reported BSS scores was moderate (ICC = 0.61). AI stool characterization also correlated better than subject reports with diarrhea severity scores.

Discussion: A novel smartphone application can determine BSS and other visual stool characteristics with high accuracy compared to 2 expert gastroenterologists. Moreover, trained AI was superior to subject self-reporting of BSS. AI assessments could provide more objective outcome measures for stool characterization in gastroenterology.