{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T15:00:21Z","timestamp":1790694021979,"version":"4.1.0"},"reference-count":39,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T00:00:00Z","timestamp":1790640000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Nutr."],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Accurate carbohydrate estimation is critical for insulin dosing in individuals with diabetes on multiple daily injections. Conventional methods such as visual estimation and food exchange systems are inconsistent, particularly in multicultural Asian settings where portion sizes vary and nutritional labeling is limited. NutriDish is an AI-powered system designed to automates carbohydrate estimation from food images.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>This study compared the accuracy of carbohydrate estimates generated by NutriDish and ChatGPT-5 against reference values established by a dietitian using weighed food portions.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Between July and August 2025, food items were collected from hawker centers, food courts, restaurants, and homes, supermarkets, and convenience stores across Singapore to capture the region's diverse culinary landscape. Items were classified as simple foods or composite meals. NutriDish and ChatGPT-5 independently estimated carbohydrate content, while a dietitian established reference values using weighing portions and a validated nutrient database. Accuracy was assessed using Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), proportion of estimates within \u00b1 10 g of the reference value, and Bland-Altman analysis.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      Across 150 food items, NutriDish achieved significantly lower MAE [5.9 \u00b1 5.3 g (median: 4.1 g, IQR: 1.0\u20139.0) vs. 8.7 \u00b1 7.4 g (median: 6.0 g, IQR: 3.0\u201313.4);\n                      <jats:italic>p<\/jats:italic>\n                      &amp;lt; 0.001] and MAPE [22.1 \u00b1 35.7% (median: 13.0%, IQR: 4.9\u201325.0) vs. 32.9 \u00b1 50.8% (median: 22.1%, IQR: 9.1\u201337.1);\n                      <jats:italic>p<\/jats:italic>\n                      &amp;lt; 0.001] than ChatGPT-5. NutriDish estimates fell within \u00b1 10 g of the reference values in 80.7% of cases vs. 69.3% for ChatGPT-5 (\n                      <jats:italic>p<\/jats:italic>\n                      = 0.016). Simple food items were estimated more accurately than composite meals, with MAEs of 4.1 \u00b1 4.3 g and 8.6 \u00b1 5.6 g, respectively (\n                      <jats:italic>p<\/jats:italic>\n                      &amp;lt; 0.001). Regarding reference object types, nutrition labels yielded the highest accuracy (MAE: 2.7 \u00b1 4.2 g; MAPE: 9.0% \u00b1 14.5%), whereas coins were associated with the poorest performance (MAE: 9.3 \u00b1 7.3 g; MAPE: 48.1% \u00b1 74.6%).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>NutriDish demonstrated good agreement with dietitian-derived carbohydrate estimates, highlighting its potential to aid carbohydrate counting in individuals treated with insulin. Future studies should evaluate real-world clinical usability, integration into diabetes self-management ecosystems and impact on long-term glycemic control.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fnut.2026.1927840","type":"journal-article","created":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T14:32:03Z","timestamp":1790692323000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Comparative accuracy of automated carbohydrate estimation using NutriDish and ChatGPT-5 against dietitian-derived reference values"],"prefix":"10.3389","volume":"13","author":[{"given":"Xia","family":"Lian","sequence":"first","affiliation":[{"name":"Department of Nursing, Tan Tock Seng Hospital","place":["Singapore, Singapore"]},{"name":"Department of Endocrinology, Tan Tock Seng Hospital","place":["Singapore, Singapore"]},{"name":"Alice Lee Center for Nursing Studies, Yong Loo Lin School of Medicine, National University of Singapore","place":["Singapore, Singapore"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Weixiang","family":"Lian","sequence":"additional","affiliation":[{"name":"Department of Epidemiology & Preventive Medicine, Tan Tock Seng Hospital","place":["Singapore, Singapore"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hwee Pin","family":"Phua","sequence":"additional","affiliation":[{"name":"Department of Epidemiology & Preventive Medicine, Tan Tock Seng Hospital","place":["Singapore, Singapore"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alvin Wai Kit","family":"Tan","sequence":"additional","affiliation":[{"name":"Department of Endocrinology, Tan Tock Seng Hospital","place":["Singapore, Singapore"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Huiling","family":"Liew","sequence":"additional","affiliation":[{"name":"Department of Endocrinology, Tan Tock Seng Hospital","place":["Singapore, Singapore"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Eunice Yu Wen","family":"Goh","sequence":"additional","affiliation":[{"name":"Department of Nursing, Tan Tock Seng Hospital","place":["Singapore, Singapore"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Christina Yang Hoon","family":"Goh","sequence":"additional","affiliation":[{"name":"Department of Nursing, Tan Tock Seng Hospital","place":["Singapore, Singapore"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Melissa Hui Li","family":"Ho","sequence":"additional","affiliation":[{"name":"Department of Nutrition & Dietetics, Tan Tock Seng Hospital","place":["Singapore, Singapore"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Puay Shi","family":"Ng","sequence":"additional","affiliation":[{"name":"Department of Nutrition & Dietetics, Tan Tock Seng Hospital","place":["Singapore, Singapore"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Isaac Jun Song","family":"Tan","sequence":"additional","affiliation":[{"name":"Department of Nursing, Tan Tock Seng Hospital","place":["Singapore, Singapore"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Helen","family":"Lim","sequence":"additional","affiliation":[{"name":"Department of Nursing, Tan Tock Seng Hospital","place":["Singapore, Singapore"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sanchalika","family":"Acharyya","sequence":"additional","affiliation":[{"name":"Clinical Research & Innovation Office, Tan Tock Seng Hospital","place":["Singapore, Singapore"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jocelyn Han Shi","family":"Chew","sequence":"additional","affiliation":[{"name":"Alice Lee Center for Nursing Studies, Yong Loo Lin School of Medicine, National University of Singapore","place":["Singapore, Singapore"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wenru","family":"Wang","sequence":"additional","affiliation":[{"name":"Alice Lee Center for Nursing Studies, Yong Loo Lin School of Medicine, National University of Singapore","place":["Singapore, Singapore"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rinkoo","family":"Dalan","sequence":"additional","affiliation":[{"name":"Department of Endocrinology, Tan Tock Seng Hospital","place":["Singapore, Singapore"]},{"name":"Lee Kong Chian School of Medicine, Nanyang Technological University","place":["Singapore, Singapore"]},{"name":"KLARIS Precision Health Research Clinic, Department of Endocrinology, Tan Tock Seng Hospital","place":["Singapore, Singapore"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,29]]},"reference":[{"key":"B1","volume-title":"IDF diabetes atlas. 10th ed","author":"International","year":"2021"},{"key":"B2","doi-asserted-by":"publisher","first-page":"14790","DOI":"10.1038\/s41598-020-71908-9","article-title":"Global, regional, and national burden and trend of diabetes in 195 countries and territories: an analysis from 1990 to 2025","volume":"10","author":"Lin","year":"2020","journal-title":"Sci Rep."},{"key":"B3","doi-asserted-by":"publisher","first-page":"977","DOI":"10.1056\/NEJM199309303291401","article-title":"The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus","volume":"329","author":"Diabetes","year":"1993","journal-title":"New England J Med"},{"key":"B4","doi-asserted-by":"publisher","first-page":"837","DOI":"10.1016\/S0140-6736(98)07019-6","article-title":"UK Prospective Diabetes Study (UKPDS) Group","volume":"352","year":"1998","journal-title":"Lancet"},{"key":"B5","doi-asserted-by":"publisher","first-page":"e21","DOI":"10.1111\/j.1464-5491.2012.03595.x","article-title":"In children using intensive insulin therapy, a 20-g variation in carbohydrate amount significantly impacts postprandial glycaemia","volume":"29","author":"Smart","year":"2012","journal-title":"Diabet Med."},{"key":"B6","doi-asserted-by":"publisher","first-page":"279","DOI":"10.1111\/j.1464-5491.2009.02669.x","article-title":"Children and adolescents on intensive insulin therapy maintain postprandial glycaemic control without precise carbohydrate counting","volume":"26","author":"Smart","year":"2009","journal-title":"Diabet Med."},{"key":"B7","doi-asserted-by":"publisher","first-page":"56","DOI":"10.2337\/diaspect.22.1.56","article-title":"The carbohydrate counting in adolescents with type 1 diabetes (CCAT) study","volume":"22","author":"Bishop","year":"2009","journal-title":"Diabet Spectr."},{"key":"B8","doi-asserted-by":"publisher","first-page":"19","DOI":"10.1016\/j.diabres.2012.10.024","article-title":"Carbohydrate counting accuracy and blood glucose variability in adults with type 1 diabetes","volume":"99","author":"Brazeau","year":"2013","journal-title":"Diabetes Res Clin Pract."},{"key":"B9","doi-asserted-by":"publisher","first-page":"348","DOI":"10.1111\/j.1464-5491.2010.02945.x","article-title":"Can children with type 1 diabetes and their caregivers estimate the carbohydrate content of meals and snacks?","volume":"27","author":"Smart","year":"2010","journal-title":"Diabet Med."},{"key":"B10","doi-asserted-by":"publisher","first-page":"238","DOI":"10.1016\/S0168-8227(00)82267-6","article-title":"How well are individuals on intensive insulin therapy counting carbohydrates?","volume":"50","author":"Graff","year":"2000","journal-title":"Diabet Res Clin Pract."},{"key":"B11","doi-asserted-by":"publisher","first-page":"214","DOI":"10.1016\/j.diabres.2017.02.022","article-title":"Practices, perceptions and expectations for carbohydrate counting in patients with type 1 diabetes - Results from an online survey","volume":"126","author":"Fortin","year":"2017","journal-title":"Diabetes Res Clin Pract."},{"key":"B12","doi-asserted-by":"publisher","first-page":"153","DOI":"10.1297\/cpe.24.153","article-title":"The factors affecting on estimation of carbohydrate content of meals in carbohydrate counting","volume":"24","author":"Kawamura","year":"2015","journal-title":"J Clin Res Pediatr Endocrinol."},{"key":"B13","doi-asserted-by":"publisher","first-page":"142","DOI":"10.2337\/diaclin.34.3.142","article-title":"Accuracy of carbohydrate counting in adults","volume":"34","author":"Meade","year":"2016","journal-title":"Clinical Diabetes."},{"key":"B14","unstructured":"Singapore\n          Health Promotion Board.\n          National Nutrition Survey 2022\n          \n          2022"},{"key":"B15","doi-asserted-by":"publisher","first-page":"e101","DOI":"10.2196\/jmir.5567","article-title":"Carbohydrate estimation by a mobile phone-based system versus self-estimations of individuals with type 1 diabetes mellitus: a comparative study","volume":"18","author":"Rhyner","year":"2016","journal-title":"J Med Internet Res."},{"key":"B16","doi-asserted-by":"publisher","first-page":"1570","DOI":"10.1177\/19322968241264744","article-title":"Carbohydrate estimation accuracy of two commercially available smartphone applications vs estimation by individuals with type 1 diabetes: a comparative study","volume":"19","author":"Baumgartner","year":"2024","journal-title":"J Diabetes Sci Technol."},{"key":"B17","doi-asserted-by":"publisher","first-page":"e15294","DOI":"10.2196\/15294","article-title":"Volumetric food quantification using computer vision on a depth-sensing smartphone: preclinical study","volume":"8","author":"Herzig","year":"2020","journal-title":"JMIR mHealth uHealth."},{"key":"B18","doi-asserted-by":"publisher","first-page":"e15602","DOI":"10.2196\/15602","article-title":"Use of different food image recognition platforms in dietary assessment: comparison study","volume":"4","author":"Asbroeck","year":"2020","journal-title":"JMIR Format Res."},{"key":"B19","doi-asserted-by":"publisher","first-page":"732449","DOI":"10.3389\/fnut.2021.732449","article-title":"Feasibility study of an automated carbohydrate estimation system using thai food images in comparison with estimation by dietitians","volume":"8","author":"Chotwanvirat","year":"2021","journal-title":"Front Nutr"},{"key":"B20","doi-asserted-by":"publisher","first-page":"19322968261424270","DOI":"10.1177\/19322968261424270","article-title":"Evaluating accuracy of chatgpt-4o in automated carbohydrate estimation from images as a self-management tool for adolescents with type 1 diabetes","volume":"28","author":"Johansen","year":"2026","journal-title":"J Diabet Sci Technol"},{"key":"B21","doi-asserted-by":"publisher","first-page":"1048","DOI":"10.1177\/19322968261431820","article-title":"Augmented food image analysis with multimodal large language models to support carbohydrate counting in diabetes","volume":"20","author":"Johansen","year":"2026","journal-title":"J Diabet Sci Technol"},{"key":"B22","doi-asserted-by":"publisher","first-page":"7577","DOI":"10.1109\/JBHI.2024.3417280","article-title":"Dietary assessment with multimodal ChatGPT: a systematic analysis","volume":"28","author":"Lo","year":"2024","journal-title":"IEEE J Biomed Health Inform."},{"key":"B23","doi-asserted-by":"publisher","first-page":"e97","DOI":"10.2337\/dc25-0303","article-title":"AI-powered carbohydrate counting for type 1 diabetes: accuracy and real-world performance","volume":"48","author":"Tecce","year":"2025","journal-title":"Diabetes Care."},{"key":"B24","doi-asserted-by":"publisher","first-page":"167","DOI":"10.4103\/ijem.ijem_369_22","article-title":"Type 1 Diabetes in Singapore: self-care challenges, diabetes technology awareness, current use, and satisfaction, an online survey","volume":"28","author":"Chandran","year":"2024","journal-title":"Indian J Endocrinol Metab."},{"key":"B25","unstructured":"Ltd\n          \n          \n            \n              Buzud\n              Pte\n            \n          \n          Buzud: Reimagined Healthcare\n          \n          2024"},{"key":"B26","unstructured":"ChatGPT"},{"key":"B27","unstructured":""},{"key":"B28","unstructured":"Singapore FoodID [database]"},{"key":"B29","unstructured":"Xyris\n              Pty Ltd\n            \n          \n          FoodWorks Online Professional\n          \n          2025"},{"key":"B30","unstructured":"Singapore\n          Health Promotion Board\n          Energy and nutrient composition of food [internet]\n          \n          2025"},{"key":"B31","unstructured":"nBuddy \u2013 Nutritionist Buddy [mobile application]\n          \n          2025"},{"key":"B32","unstructured":"R\n              Core Team\n            \n          \n          R: A language and environment for statistical computing\n          \n          2025"},{"key":"B33","doi-asserted-by":"publisher","first-page":"753","DOI":"10.1177\/1932296816679850","article-title":"Accurate carbohydrate counting is an important determinant of postprandial glycemia in children and adolescents with type 1 diabetes on insulin pump therapy","volume":"11","author":"Deeb","year":"2017","journal-title":"J Diabetes Sci Technol."},{"key":"B34","doi-asserted-by":"publisher","first-page":"401","DOI":"10.1016\/j.arcontrol.2019.06.003","article-title":"Advanced carbohydrate counting: an engineering perspective","volume":"48","author":"Reiterer","year":"2019","journal-title":"Annu Rev Control."},{"key":"B35","doi-asserted-by":"publisher","first-page":"227","DOI":"10.1111\/j.1464-5491.2010.03176.x","article-title":"Effects of meals with different glycaemic index on postprandial blood glucose response in patients with Type 1 diabetes treated with continuous subcutaneous insulin infusion","volume":"28","author":"Parillo","year":"2011","journal-title":"Diabet Med."},{"key":"B36","doi-asserted-by":"publisher","first-page":"173","DOI":"10.1089\/dia.2009.0112","article-title":"Timing of meal insulin boluses to achieve optimal postprandial glycemic control in patients with type 1 diabetes","volume":"12","author":"Cobry","year":"2010","journal-title":"Diabetes Technol Ther."},{"key":"B37","doi-asserted-by":"publisher","first-page":"149","DOI":"10.2337\/ds20-0004","article-title":"Factors beyond carbohydrate to consider when determining meantime insulin doses: protein, fat, timing, and technology","volume":"33","author":"Evert","year":"2020","journal-title":"Diabetes Spectr."},{"key":"B38","doi-asserted-by":"publisher","first-page":"79","DOI":"10.1007\/s00125-021-05587-0","article-title":"Dietary determinants of postprandial blood glucose control in adults with type 1 diabetes on a hybrid closed-loop system","volume":"65","author":"Vetrani","year":"2022","journal-title":"Diabetologia."},{"key":"B39","doi-asserted-by":"publisher","first-page":"6","DOI":"10.1186\/1472-6823-12-6","article-title":"Assessment of endogenous insulin secretion in insulin-treated diabetes predicts postprandial glucose and treatment response to prandial insulin","volume":"12","author":"Jones","year":"2012","journal-title":"BMC Endocr Disord."}],"container-title":["Frontiers in Nutrition"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnut.2026.1927840\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T14:32:08Z","timestamp":1790692328000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnut.2026.1927840\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,29]]},"references-count":39,"alternative-id":["10.3389\/fnut.2026.1927840"],"URL":"https:\/\/doi.org\/10.3389\/fnut.2026.1927840","relation":{},"ISSN":["2296-861X"],"issn-type":[{"value":"2296-861X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,29]]},"article-number":"1927840"}}