{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T02:05:08Z","timestamp":1789610708596,"version":"build-2803163510"},"reference-count":25,"publisher":"Ovid Technologies (Wolters Kluwer Health)","issue":"10","content-domain":{"domain":["lww.com","ovid.com"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2026,10]]},"abstract":"<jats:sec>\n                    <jats:title>BACKGROUND:<\/jats:title>\n                    <jats:p>Surgeons and ostomy nurses receive a high volume of stoma photos from patients.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>OBJECTIVE:<\/jats:title>\n                    <jats:p>This study aimed to develop and validate an automated and scalable artificial intelligence pipeline for identification and triage of patient-submitted intestinal ostomy photos.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>DESIGN:<\/jats:title>\n                    <jats:p>This was a retrospective derivation and validation study with stakeholder engagement to guide model design. Clinical teams reviewed photos in duplicate and categorized them into: 1) obstructed view, 2) healthy stoma, 3) suitable for conservative management, or 4) requiring in-person review. Pretrained neural networks, including MobileNetV4, ResNet50, vision transformer, and contrastive language-image pretraining vision transformer, were fine-tuned with 5-fold cross validation.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>SETTINGS:<\/jats:title>\n                    <jats:p>Nine Mayo Clinic hospitals (2019\u20132022).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>PATIENTS:<\/jats:title>\n                    <jats:p>Adult patients (age 18 years or older) undergoing surgery who submitted images within 30 days after surgery were included in the study.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>MAIN OUTCOME MEASURES:<\/jats:title>\n                    <jats:p>Model performance was evaluated with area under the receiver operating characteristic curve, precision, recall, and F1 scores.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>RESULTS:<\/jats:title>\n                    <jats:p>In this study, 538 photos of abdominal ostomies were sent in by 191 patients (median age 52 years; interquartile range, 40\u201363; 60.4% women). Expert consensus triaged 236 photos (43.9%) as having an obstructed view, 30 photos (5.6%) as healthy, 177 photos (32.9%) for conservative management, and 95 photos (17.7%) for in-person review. All models achieved an area under the receiver operating characteristic curve greater than 0.97 for identifying stomas. The contrastive language-image pretraining vision transformer model performed best at triage (macro-area under the receiver operating characteristic curve 0.94\u2009\u00b1\u20090.06; F1 score 0.77\u2009\u00b1\u20090.10). End-to-end detection and triage using contrastive language-image pretraining vision transformer achieved an area under the receiver operating characteristic curve of 0.99\u2009\u00b1\u20090.01, precision of 0.87\u2009\u00b1\u20090.08, recall of 0.93\u2009\u00b1\u20090.04, and F1 score of 0.89\u2009\u00b1\u20090.07. Attention maps showed that the models focused on stomas to determine classification.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>LIMITATIONS:<\/jats:title>\n                    <jats:p>Low sample size and lack of prospective validation.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>CONCLUSIONS:<\/jats:title>\n                    <jats:p>\n                      A vision-language model pipeline accurately detected and triaged patient-submitted ostomy photos. Prospective evaluation is now needed to support integration into multidisciplinary digital workflows. See\n                      <jats:bold>Video Abstract<\/jats:bold>\n                      .\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>ENFOQUE DE VISI\u00d3N ARTIFICIAL PARA LA CLASIFICACI\u00d3N DE FOTOGRAF\u00cdAS DE OSTOM\u00cdAS INTESTINALES ENVIADAS POR LOS PACIENTES<\/jats:title>\n                    <jats:p>\n                      <jats:bold>ANTECEDENTES:<\/jats:bold>\n                    <\/jats:p>\n                    <jats:p>Los cirujanos y el personal de enfermer\u00eda especializado en ostom\u00edas reciben un gran volumen de fotograf\u00edas de estomas de los pacientes.<\/jats:p>\n                    <jats:p>\n                      <jats:bold>OBJETIVO:<\/jats:bold>\n                    <\/jats:p>\n                    <jats:p>Este estudio tuvo como objetivo desarrollar y validar un sistema automatizado y escalable de inteligencia artificial para la identificaci\u00f3n y clasificaci\u00f3n de fotograf\u00edas de ostom\u00edas intestinales enviadas por los pacientes.<\/jats:p>\n                    <jats:p>\n                      <jats:bold>DISE\u00d1O:<\/jats:bold>\n                    <\/jats:p>\n                    <jats:p>Estudio retrospectivo de derivaci\u00f3n y validaci\u00f3n con la participaci\u00f3n de las partes interesadas para guiar el dise\u00f1o del modelo. Los equipos cl\u00ednicos revisaron las fotograf\u00edas por duplicado y las clasificaron en: (a) visi\u00f3n obstruida, (b) estoma sano, (c) apto para tratamiento conservador o (d) que requiere revisi\u00f3n presencial. Se optimizaron redes neuronales preentrenadas, incluyendo MobileNetV4, ResNet50, ViT y CLIP-ViT, mediante validaci\u00f3n cruzada de 5 pliegues.<\/jats:p>\n                    <jats:p>\n                      <jats:bold>\u00c1MBITO:<\/jats:bold>\n                    <\/jats:p>\n                    <jats:p>Nueve hospitales de la Cl\u00ednica Mayo (2019-2022).<\/jats:p>\n                    <jats:p>\n                      <jats:bold>PACIENTES:<\/jats:bold>\n                    <\/jats:p>\n                    <jats:p>Pacientes adultos (\u226518 a\u00f1os) sometidos a cirug\u00eda que enviaron im\u00e1genes dentro de los 30 d\u00edas posteriores a la intervenci\u00f3n.<\/jats:p>\n                    <jats:p>\n                      <jats:bold>PRINCIPALES MEDIDAS DE RESULTADO:<\/jats:bold>\n                    <\/jats:p>\n                    <jats:p>El rendimiento del modelo se evalu\u00f3 mediante el \u00e1rea bajo la curva ROC, la precisi\u00f3n, la exhaustividad y la puntuaci\u00f3n F1.<\/jats:p>\n                    <jats:p>\n                      <jats:bold>RESULTADOS:<\/jats:bold>\n                    <\/jats:p>\n                    <jats:p>En este estudio, 538 fotograf\u00edas de ostom\u00edas abdominales fueron enviadas por 191 pacientes (edad media 52 a\u00f1os, rango intercuart\u00edlico 40-63; 60,4 % mujeres). El consenso de expertos clasific\u00f3 236 (43,9 %) fotograf\u00edas como con visi\u00f3n obstruida, 30 (5,6 %) como sanas, 177 (32,9 %) para tratamiento conservador y 95 (17,7 %) para revisi\u00f3n presencial. Todos los modelos lograron un \u00e1rea bajo la curva ROC &gt;0,97 para la identificaci\u00f3n de estomas. El modelo CLIP-ViT tuvo el mejor rendimiento en la clasificaci\u00f3n (macro: \u00e1rea bajo la curva ROC 0,94\u2009\u00b1\u20090,06; F1 0,77\u2009\u00b1\u20090,10). La detecci\u00f3n y clasificaci\u00f3n de extremo a extremo mediante CLIP-ViT alcanz\u00f3 un \u00e1rea bajo la curva ROC de 0,99\u2009\u00b1\u20090,01, una precisi\u00f3n de 0,87\u2009\u00b1\u20090,08, una exhaustividad de 0,93\u2009\u00b1\u20090,04 y una puntuaci\u00f3n F1 de 0,89\u2009\u00b1\u20090,07. Los mapas de atenci\u00f3n mostraron que los modelos se centraron en los estomas para determinar la clasificaci\u00f3n.<\/jats:p>\n                    <jats:p>\n                      <jats:bold>LIMITACIONES:<\/jats:bold>\n                    <\/jats:p>\n                    <jats:p>Tama\u00f1o de muestra reducido y falta de validaci\u00f3n prospectiva.<\/jats:p>\n                    <jats:p>\n                      <jats:bold>CONCLUSIONES:<\/jats:bold>\n                    <\/jats:p>\n                    <jats:p>\n                      Un modelo de lenguaje visual detect\u00f3 y clasific\u00f3 con precisi\u00f3n las fotos de ostom\u00edas enviadas por los pacientes. Ahora se requiere una evaluaci\u00f3n prospectiva para respaldar su integraci\u00f3n en flujos de trabajo digitales multidisciplinarios.\n                      <jats:italic toggle=\"yes\">(AI-generated translation<\/jats:italic>\n                      )\n                    <\/jats:p>\n                  <\/jats:sec>","DOI":"10.1097\/dcr.0000000000004366","type":"journal-article","created":{"date-parts":[[2026,7,17]],"date-time":"2026-07-17T13:27:31Z","timestamp":1784294851000},"page":"2521-2530","update-policy":"https:\/\/doi.org\/10.1097\/lww.0000000000001000","source":"Crossref","is-referenced-by-count":0,"title":["Computer Vision Approach to Triaging Patient-Submitted Photos of Intestinal Ostomies"],"prefix":"10.1097","volume":"69","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7369-8639","authenticated-orcid":false,"given":"Chris","family":"Varghese","sequence":"first","affiliation":[{"name":"Division of Hepatobiliary and Pancreas Surgery, Mayo Clinic, Rochester, Minnesota"},{"name":"Department of Surgery, University of Auckland, Auckland, New Zealand"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ashok","family":"Choudhary","sequence":"additional","affiliation":[{"name":"Division of Hepatobiliary and Pancreas Surgery, Mayo Clinic, Rochester, Minnesota"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ellen L.","family":"Larson","sequence":"additional","affiliation":[{"name":"Division of Hepatobiliary and Pancreas Surgery, Mayo Clinic, Rochester, Minnesota"},{"name":"Division of Colon and Rectal Surgery, Mayo Clinic, Rochester, Minnesota"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Frank G.","family":"Lee","sequence":"additional","affiliation":[{"name":"Division of Hepatobiliary and Pancreas Surgery, Mayo Clinic, Rochester, Minnesota"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Michele C.","family":"Baier","sequence":"additional","affiliation":[{"name":"Division of Colon and Rectal Surgery, Mayo Clinic, Rochester, Minnesota"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Cody L.","family":"Glabe","sequence":"additional","affiliation":[{"name":"Division of Colon and Rectal Surgery, Mayo Clinic, Rochester, Minnesota"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jora N.","family":"Just","sequence":"additional","affiliation":[{"name":"Division of Colon and Rectal Surgery, Mayo Clinic, Rochester, Minnesota"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Stacy A.","family":"Friedrich","sequence":"additional","affiliation":[{"name":"Division of Colon and Rectal Surgery, Mayo Clinic, Rochester, Minnesota"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Amit","family":"Merchea","sequence":"additional","affiliation":[{"name":"Division of Colon and Rectal Surgery, Mayo Clinic, Jacksonville, Florida"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Justin T.","family":"Brady","sequence":"additional","affiliation":[{"name":"Division of Colon and Rectal Surgery, Mayo Clinic, Phoenix, Arizona"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jody L.","family":"Selig","sequence":"additional","affiliation":[{"name":"Division of Colon and Rectal Surgery, Mayo Clinic, Rochester, Minnesota"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Cornelius A.","family":"Thiels","sequence":"additional","affiliation":[{"name":"Division of Hepatobiliary and Pancreas Surgery, Mayo Clinic, Rochester, Minnesota"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hojjat","family":"Salehinejad","sequence":"additional","affiliation":[{"name":"Robert D. and Patricia E. Kern Center for the Science of Health Care Delivery, Mayo Clinic, Rochester, Minnesota"},{"name":"Department of Artificial Intelligence and Informatics, Mayo Clinic, Rochester, Minnesota"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8718-8306","authenticated-orcid":false,"given":"William R.G.","family":"Perry","sequence":"additional","affiliation":[{"name":"Division of Colon and Rectal Surgery, Mayo Clinic, Rochester, Minnesota"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"276","published-online":{"date-parts":[[2026,7,17]]},"reference":[{"key":"R2-20260917","doi-asserted-by":"crossref","first-page":"2286","DOI":"10.1111\/codi.15788","article-title":"Long-term outcome and quality of life after continent ileostomy for ulcerative colitis: a systematic review.","volume":"23","author":"Deputy","year":"2021","journal-title":"Colorectal Dis"},{"key":"R3-20260917","doi-asserted-by":"crossref","first-page":"51","DOI":"10.1097\/DCR.0000000000000732","article-title":"Long-term quality of life of patients with permanent end ileostomy: results of a nationwide cross-sectional survey.","volume":"60","author":"Schiergens","year":"2017","journal-title":"Dis Colon Rectum"},{"key":"R4-20260917","doi-asserted-by":"crossref","first-page":"861","DOI":"10.1111\/codi.16467","article-title":"Characterising nationwide reasons for unplanned hospital readmission after colorectal cancer surgery.","volume":"25","author":"Xu","year":"2023","journal-title":"Colorectal Dis"},{"key":"R5-20260917","doi-asserted-by":"crossref","first-page":"1071","DOI":"10.1111\/codi.15566","article-title":"Risk factors for readmission with dehydration after ileostomy formation: a systematic review and meta-analysis.","volume":"23","author":"Liu","year":"2021","journal-title":"Colorectal Dis"},{"key":"R6-20260917","doi-asserted-by":"crossref","first-page":"1852","DOI":"10.1111\/codi.16658","article-title":"Patient satisfaction with stoma care and their expectations on mobile apps for supportive care.","volume":"25","author":"van der Storm","year":"2023","journal-title":"Colorectal Dis"},{"key":"R7-20260917","doi-asserted-by":"crossref","first-page":"1820","DOI":"10.1111\/ans.70269","article-title":"Follow-up strategies in colorectal cancer: what do patients prefer?","volume":"95","author":"Srikumar","year":"2025","journal-title":"ANZ J Surg"},{"key":"R8-20260917","doi-asserted-by":"crossref","first-page":"139","DOI":"10.1007\/s13304-023-01671-9","article-title":"Digital postoperative follow-up after colorectal resection: a multi-center preliminary qualitative study on a patient reporting and monitoring application.","volume":"76","author":"Bertoni","year":"2024","journal-title":"Updates Surg"},{"key":"R9-20260917","doi-asserted-by":"crossref","first-page":"1304","DOI":"10.1093\/bjs\/znab323","article-title":"Mobile health technology for remote home monitoring after surgery: a meta-analysis.","volume":"108","author":"Dawes","year":"2021","journal-title":"Br J Surg"},{"key":"R10-20260917","doi-asserted-by":"crossref","first-page":"e295","DOI":"10.1016\/S2589-7500(23)00026-2","article-title":"Readiness for implementation of novel digital health interventions for postoperative monitoring: a systematic review and clinical innovation network analysis.","volume":"5","author":"McLean","year":"2023","journal-title":"Lancet Digit Health"},{"key":"R11-20260917","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1038\/s41746-021-00525-1","article-title":"Mobile devices and wearable technology for measuring patient outcomes after surgery: a systematic review.","volume":"4","author":"Knight","year":"2021","journal-title":"NPJ. Digit Med"},{"key":"R12-20260917","doi-asserted-by":"crossref","first-page":"559","DOI":"10.1001\/jamainternmed.2016.0209","article-title":"The burden of inbox notifications in commercial electronic health records.","volume":"176","author":"Murphy","year":"2016","journal-title":"JAMA Intern Med"},{"key":"R13-20260917","doi-asserted-by":"crossref","first-page":"361","DOI":"10.1016\/j.jaad.2017.12.026","article-title":"The burden of inbox-messaging systems and its effect on work-life balance in dermatology.","volume":"79","author":"Bittar","year":"2018","journal-title":"J Am Acad Dermatol"},{"key":"R14-20260917","doi-asserted-by":"crossref","first-page":"492","DOI":"10.1016\/j.jaad.2021.09.026","article-title":"The disproportionate burden of electronic health record messages with image attachments in dermatology.","volume":"86","author":"Borre","year":"2022","journal-title":"J Am Acad Dermatol"},{"key":"R15-20260917","doi-asserted-by":"crossref","first-page":"85","DOI":"10.1038\/s41746-023-00824-9","article-title":"Evaluation of remote digital postoperative wound monitoring in routine surgical practice.","volume":"6","author":"McLean","year":"2023","journal-title":"NPJ Digit Med"},{"key":"R16-20260917","doi-asserted-by":"crossref","first-page":"160","DOI":"10.1038\/s41746-021-00526-0","article-title":"Remote diagnosis of surgical-site infection using a mobile digital intervention: a randomised controlled trial in emergency surgery patients.","volume":"4","author":"McLean","year":"2021","journal-title":"NPJ Digit Med"},{"key":"R17-20260917","doi-asserted-by":"crossref","first-page":"2273","DOI":"10.1111\/ans.70141","article-title":"Remote follow-up after colorectal cancer surgery: a systematic review.","volume":"95","author":"Srikumar","year":"2025","journal-title":"ANZ J Surg"},{"key":"R18-20260917","doi-asserted-by":"crossref","first-page":"1257","DOI":"10.1038\/s41591-024-02970-3","article-title":"Artificial intelligence in surgery.","volume":"30","author":"Varghese","year":"2024","journal-title":"Nat Med"},{"key":"R19-20260917","doi-asserted-by":"crossref","first-page":"121","DOI":"10.1038\/s41746-024-01419-8","article-title":"Multimodal machine learning to predict surgical site infection with healthcare workload impact assessment.","volume":"8","author":"McLean","year":"2025","journal-title":"NPJ Digit Med"},{"key":"R20-20260917","doi-asserted-by":"crossref","first-page":"419","DOI":"10.1097\/SLA.0000000000006826","article-title":"Imaging-based surgical site infection detection using artificial intelligence.","volume":"282","author":"Muaddi","year":"2025","journal-title":"Ann Surg"},{"key":"R21-20260917","doi-asserted-by":"crossref","first-page":"e078378","DOI":"10.1136\/bmj-2023-078378","article-title":"TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods.","volume":"385","author":"Collings","year":"2024","journal-title":"BMJ"},{"key":"R22-20260917","article-title":"Learning transferable visual models from natural language supervision.","author":"Radford","year":"2021","journal-title":"arXiv:2103.00020v1 [cs.CV]"},{"key":"R23-20260917","article-title":"An image is worth 16x16 words: transformers for image recognition at scale.","author":"Dosovitskiy","year":"2021","journal-title":"arXiv:2010.11929v2"},{"key":"R26-20260917","doi-asserted-by":"crossref","first-page":"e078276","DOI":"10.1136\/bmj-2023-078276","article-title":"Developing clinical prediction models: a step-by-step guide.","volume":"386","author":"Efthimiou","year":"2024","journal-title":"BMJ"},{"key":"R27-20260917","article-title":"SCLIP: rethinking self-attention for dense vision-language inference.","author":"Wang","year":"2024","journal-title":"arXiv [cs.CV]"},{"key":"R28-20260917","doi-asserted-by":"crossref","first-page":"100916","DOI":"10.1016\/j.landig.2025.100916","article-title":"Evaluation of performance measures in predictive artificial intelligence models to support medical decisions: overview and guidance.","volume":"7","author":"Van Calster","year":"2025","journal-title":"Lancet Digit Health"}],"container-title":["Diseases of the Colon &amp; Rectum"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.ovid.com\/10.1097\/DCR.0000000000004366","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T02:00:22Z","timestamp":1789610422000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.ovid.com\/10.1097\/DCR.0000000000004366"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,17]]},"references-count":25,"journal-issue":{"issue":"10","published-print":{"date-parts":[[2026]]}},"URL":"https:\/\/doi.org\/10.1097\/dcr.0000000000004366","relation":{},"ISSN":["0012-3706","1530-0358"],"issn-type":[{"value":"0012-3706","type":"print"},{"value":"1530-0358","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,17]]}}}