{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,28]],"date-time":"2026-08-28T13:32:32Z","timestamp":1787923952014,"version":"build-2784847793"},"reference-count":51,"publisher":"MDPI AG","issue":"9","license":[{"start":{"date-parts":[[2026,8,28]],"date-time":"2026-08-28T00:00:00Z","timestamp":1787875200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Bioengineering"],"abstract":"<jats:p>Background: Burn extent guides triage, transfer and fluid resuscitation, yet its clinical estimation is imprecise and observer-dependent. Multimodal large language models (MLLMs) process clinical photographs without task-specific training, but their error has rarely been separated into systematic and random components or their performance across skin tones characterized. Methods: Three state-of-the-art MLLMs (Gemini 3.1 Pro, GPT-5.6 Sol, and Fable 5) each assessed 153 burn photographs five times under an identical prompt. The tasks were as follows: burned proportion of the imaged field, against an expert-guided pixel-wise segmentation (tolerance \u00b1 10 percentage points, pp); burned percentage of total body surface area (TBSA), against physician consensus (\u00b12 pp); and binary Fitzpatrick skin tone (FST; light I\u2013III versus dark IV\u2013VI). The first of these was the primary endpoint. Results: The primary endpoint was in the range of 32.5\u201370.2%, with TBSA at 69.5\u201377.5%. All models compressed the estimation range (slopes 0.58\u20130.76, intercepts +10.0 to +22.2 pp); one multiplicative constant per model brought errors differing more than twofold into a 1.4 pp range. Across repeated queries, the median within-image range was 5.0\u201325.0 pp; averaging the five answers reduced error by only 0.24\u20132.22 pp. FST accuracy was 83.8\u201391.9% against a majority-class baseline of 81.0%. Conclusions: Averaging repeated answers removes only the smaller, random component; the larger, systematic one persists and requires calibration against reference data before clinical use can be considered.<\/jats:p>","DOI":"10.3390\/bioengineering13091000","type":"journal-article","created":{"date-parts":[[2026,8,28]],"date-time":"2026-08-28T12:27:46Z","timestamp":1787920066000},"page":"1000","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Burn Extent and Fitzpatrick Skin Tone Assessment from Clinical Photographs: Systematic and Random Error in Multimodal Large Language Models"],"prefix":"10.3390","volume":"13","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6395-6670","authenticated-orcid":false,"given":"Ibrahim","family":"G\u00fcler","sequence":"first","affiliation":[{"name":"Department of Plastic, Aesthetic and Hand Surgery, Otto-von-Guericke University, 39120 Magdeburg, Germany"},{"name":"Department of Health Management, Friedrich-Alexander-Universit\u00e4t Erlangen-N\u00fcrnberg, 90403 Nuremberg, Germany"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"conceptualization","vocabulary":"credit"},{"role":"methodology","vocabulary":"credit"},{"role":"software","vocabulary":"credit"},{"role":"validation","vocabulary":"credit"},{"role":"formal-analysis","vocabulary":"credit"},{"role":"investigation","vocabulary":"credit"},{"role":"resources","vocabulary":"credit"},{"role":"data-curation","vocabulary":"credit"},{"role":"writing-original-draft","vocabulary":"credit"},{"role":"writing-review-editing","vocabulary":"credit"},{"role":"visualization","vocabulary":"credit"},{"role":"project-administration","vocabulary":"credit"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6557-2163","authenticated-orcid":false,"given":"Armin","family":"Kraus","sequence":"additional","affiliation":[{"name":"Department of Plastic, Aesthetic and Hand Surgery, Otto-von-Guericke University, 39120 Magdeburg, Germany"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"resources","vocabulary":"credit"},{"role":"writing-review-editing","vocabulary":"credit"},{"role":"supervision","vocabulary":"credit"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7302-210X","authenticated-orcid":false,"given":"Gerrit","family":"Grieb","sequence":"additional","affiliation":[{"name":"Department of Plastic Surgery and Hand Surgery, Gemeinschaftskrankenhaus Havelh\u00f6he, 14089 Berlin, Germany"},{"name":"Department of Plastic Surgery and Hand Surgery, Medical Faculty, Rheinisch-Westf\u00e4lische Technische Hochschule Aachen, 52074 Aachen, Germany"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"resources","vocabulary":"credit"},{"role":"writing-review-editing","vocabulary":"credit"},{"role":"supervision","vocabulary":"credit"}]},{"given":"Henrik","family":"Stelling","sequence":"additional","affiliation":[{"name":"Department of Health Management, Friedrich-Alexander-Universit\u00e4t Erlangen-N\u00fcrnberg, 90403 Nuremberg, Germany"},{"name":"Practices for Nuclear Medicine, 12157 Berlin, Germany"}],"role":[{"vocabulary":"crossref","role":"author"},{"role":"conceptualization","vocabulary":"credit"},{"role":"methodology","vocabulary":"credit"},{"role":"software","vocabulary":"credit"},{"role":"validation","vocabulary":"credit"},{"role":"formal-analysis","vocabulary":"credit"},{"role":"investigation","vocabulary":"credit"},{"role":"resources","vocabulary":"credit"},{"role":"data-curation","vocabulary":"credit"},{"role":"writing-original-draft","vocabulary":"credit"},{"role":"writing-review-editing","vocabulary":"credit"},{"role":"visualization","vocabulary":"credit"},{"role":"project-administration","vocabulary":"credit"}]}],"member":"1968","published-online":{"date-parts":[[2026,8,28]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"493","DOI":"10.1016\/j.burns.2022.05.007","article-title":"Agreement of Clinical Assessment of Burn Size and Burn Depth between Referring Hospitals and Burn Centres: A Systematic Review","volume":"49","author":"Brekke","year":"2023","journal-title":"Burns"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"156","DOI":"10.1016\/S0305-4179(99)00047-9","article-title":"The Inter-Rater Reliability of Estimating the Size of Burns from Various Burn Area Chart Drawings","volume":"26","author":"Wachtel","year":"2000","journal-title":"Burns"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"56","DOI":"10.1177\/000313481508100131","article-title":"Total Body Surface Area Overestimation at Referring Institutions in Children Transferred to a Burn Center","volume":"81","author":"Swords","year":"2015","journal-title":"Am. Surg."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"241","DOI":"10.1016\/j.burns.2013.06.019","article-title":"The Potential Impact of Wrong TBSA Estimations on Fluid Resuscitation in Patients Suffering from Burns: Things to Keep in Mind","volume":"40","author":"Parvizi","year":"2014","journal-title":"Burns"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"721","DOI":"10.1093\/jbcr\/iraa212","article-title":"Interrater Agreement and Reliability of Burn Size Estimations Between Emergency Physicians and Burn Unit","volume":"42","author":"Yoo","year":"2021","journal-title":"J. Burn Care Res."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"501","DOI":"10.1016\/S0140-6736(51)91975-7","article-title":"The Exposure Treatment of Burns","volume":"1","author":"Wallace","year":"1951","journal-title":"Lancet"},{"key":"ref_7","first-page":"352","article-title":"The Estimation of Areas of Burns","volume":"79","author":"Lund","year":"1944","journal-title":"Surg. Gynecol. Obstet."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"47","DOI":"10.1016\/j.burnso.2023.12.006","article-title":"Burn Size Estimation: A Remarkable History with Clinical Practice Implications","volume":"8","author":"Carrougher","year":"2024","journal-title":"Burns Open"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"107524","DOI":"10.1016\/j.burns.2025.107524","article-title":"High-Precision Computer-Assisted Surface Area Estimation in Large Surface Burn Patients Using ImageJ-A Comparative Analysis","volume":"51","author":"Bucher","year":"2025","journal-title":"Burns"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Chang, C.W., Wang, H., Lai, F., Christian, M., Chen Huang, S., and Yi Tsai, H. (2025). Comparison of 3D and 2D Area Measurement of Acute Burn Wounds with LiDAR Technique and Deep Learning Model. Front. Artif. Intell., 8.","DOI":"10.3389\/frai.2025.1510905"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"233","DOI":"10.1016\/S0305-4179(00)00118-2","article-title":"An Audit of the Use of Laser Doppler Imaging (LDI) in the Assessment of Burns of Intermediate Depth","volume":"27","author":"Pape","year":"2001","journal-title":"Burns"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"1039","DOI":"10.1016\/j.burns.2022.07.006","article-title":"Application of Multiple Deep Learning Models for Automatic Burn Wound Assessment","volume":"49","author":"Chang","year":"2023","journal-title":"Burns"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"6","DOI":"10.1186\/s41038-018-0137-9","article-title":"Burn Image Segmentation Based on Mask Regions with Convolutional Neural Network Deep Learning Framework: More Accurate and More Convenient","volume":"7","author":"Jiao","year":"2019","journal-title":"Burns Trauma"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"1586","DOI":"10.1016\/j.burns.2021.01.011","article-title":"Improving Burn Depth Assessment for Pediatric Scalds by AI Based on Semantic Segmentation of Polarized Light Photography Images","volume":"47","author":"Cirillo","year":"2021","journal-title":"Burns"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"6","DOI":"10.1093\/jbcr\/irae121","article-title":"Comparing Artificial Intelligence Guided Image Assessment to Current Methods of Burn Assessment","volume":"46","author":"Lee","year":"2025","journal-title":"J. Burn Care Res."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"281","DOI":"10.3390\/ebj2040020","article-title":"Accuracy of Image-Based Automated Diagnosis in the Identification and Classification of Acute Burn Injuries. A Systematic Review","volume":"2","author":"Boissin","year":"2021","journal-title":"Eur. Burn J."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"1794","DOI":"10.1038\/s41598-023-28164-4","article-title":"Development and Evaluation of Deep Learning Algorithms for Assessment of Acute Burns and the Need for Surgery","volume":"13","author":"Boissin","year":"2023","journal-title":"Sci. Rep."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"e101102","DOI":"10.1136\/bmjhci-2024-101102","article-title":"Generative Artificial Intelligence in Primary Care: An Online Survey of UK General Practitioners","volume":"31","author":"Blease","year":"2024","journal-title":"BMJ Health Care Inform."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"e64290","DOI":"10.2196\/64290","article-title":"Laypeople\u2019s Use of and Attitudes Toward Large Language Models and Search Engines for Health Queries: Survey Study","volume":"27","author":"Mendel","year":"2025","journal-title":"J. Med. Internet Res."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"54","DOI":"10.1186\/s13049-026-01577-6","article-title":"Multimodal Large Language Model versus Emergency Physicians for Burn Assessment: A Prospective Non-Inferiority Study","volume":"34","author":"Aykut","year":"2026","journal-title":"Scand. J. Trauma Resusc. Emerg. Med."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"e68366","DOI":"10.2196\/68366","article-title":"AI-Driven Integrated System for Burn Depth Prediction with Electronic Medical Records: Algorithm Development and Validation","volume":"13","author":"Rahman","year":"2025","journal-title":"JMIR Med. Inform."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"1286","DOI":"10.1007\/s11606-025-09888-7","article-title":"Variation in Large Language Model Recommendations in Challenging Inpatient Management Scenarios","volume":"41","author":"Landon","year":"2026","journal-title":"J. Gen. Intern. Med."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"e2440969","DOI":"10.1001\/jamanetworkopen.2024.40969","article-title":"Large Language Model Influence on Diagnostic Reasoning: A Randomized Clinical Trial","volume":"7","author":"Goh","year":"2024","journal-title":"JAMA Netw. Open"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"609","DOI":"10.1038\/s41591-025-04074-y","article-title":"Reliability of LLMs as Medical Assistants for the General Public: A Randomized Preregistered Study","volume":"32","author":"Bean","year":"2026","journal-title":"Nat. Med."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"e84120","DOI":"10.2196\/84120","article-title":"Knowledge-Practice Performance Gap in Clinical Large Language Models: Systematic Review of 39 Benchmarks","volume":"27","author":"Gong","year":"2025","journal-title":"J. Med. Internet Res."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"869","DOI":"10.1001\/archderm.1988.01670060015008","article-title":"The Validity and Practicality of Sun-Reactive Skin Types I through VI","volume":"124","author":"Fitzpatrick","year":"1988","journal-title":"Arch. Dermatol."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"eabq6147","DOI":"10.1126\/sciadv.abq6147","article-title":"Disparities in Dermatology AI Performance on a Diverse, Curated Clinical Image Set","volume":"8","author":"Daneshjou","year":"2022","journal-title":"Sci. Adv."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"e240300","DOI":"10.1148\/ryai.240300","article-title":"Checklist for Artificial Intelligence in Medical Imaging (CLAIM): 2024 Update","volume":"6","author":"Tejani","year":"2024","journal-title":"Radiol. Artif. Intell."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"3283","DOI":"10.1038\/s41591-025-03953-8","article-title":"The STARD-AI Reporting Guideline for Diagnostic Accuracy Studies Using Artificial Intelligence","volume":"31","author":"Sounderajah","year":"2025","journal-title":"Nat. Med."},{"key":"ref_30","unstructured":"VEDAs Lab (2026, July 27). EBIS: Extended Burn Image Segmentation Dataset and BuRnSNet Model. Available online: https:\/\/github.com\/VEDAs-Lab\/EBIS."},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Chauhan, J., and Goyal, P. (2020, January 19\u201324). Deep Learning Based Fully Automatic Efficient Burn Severity Estimators for Better Burn Diagnosis. Proceedings of the International Joint Conference on Neural Networks (IJCNN), Glasgow, UK.","DOI":"10.1109\/IJCNN48605.2020.9206646"},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"1407","DOI":"10.1016\/j.burns.2020.03.007","article-title":"BPBSAM: Body Part-Specific Burn Severity Assessment Model","volume":"46","author":"Chauhan","year":"2020","journal-title":"Burns"},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"854","DOI":"10.1016\/j.burns.2020.08.016","article-title":"Convolution Neural Network for Effective Burn Region Segmentation of Color Images","volume":"47","author":"Chauhan","year":"2021","journal-title":"Burns"},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Chauhan, J., Rosin, P.L., and Goyal, P. (2024, January 27\u201330). BuRnSNet: Burn Region Segmentation Network from Color Images with Two-Way CNN. Proceedings of the IEEE International Conference on Image Processing, Abu Dhabi, United Arab Emirates.","DOI":"10.1109\/ICIP51287.2024.10647789"},{"key":"ref_35","unstructured":"Chauhan, J., and Singh, P.P. (October, January 27). Learning to Segment Burns with Limited Labels: A Dual-Path Cascaded Network with Uncertainty-Guided Teacher Diffusion Exposure. Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), Strasbourg, France."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"191","DOI":"10.1038\/s41746-024-01176-8","article-title":"A Survey of Skin Tone Assessment in Prospective Research","volume":"7","author":"Weir","year":"2024","journal-title":"npj Digit. Med."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"e5866","DOI":"10.1097\/GOX.0000000000005866","article-title":"Considerations for the Use of Fitzpatrick Skin Type in Plastic Surgery Research","volume":"12","author":"Subedi","year":"2024","journal-title":"Plast. Reconstr. Surg. Glob. Open"},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"30","DOI":"10.1037\/1082-989X.1.1.30","article-title":"Forming Inferences About Some Intraclass Correlation Coefficients","volume":"1","author":"Mcgraw","year":"1996","journal-title":"Psychol. Methods"},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"155","DOI":"10.1016\/j.jcm.2016.02.012","article-title":"A Guideline of Selecting and Reporting Intraclass Correlation Coefficients for Reliability Research","volume":"15","author":"Koo","year":"2016","journal-title":"J. Chiropr. Med."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"307","DOI":"10.1016\/S0140-6736(86)90837-8","article-title":"Statistical Methods for Assessing Agreement between Two Methods of Clinical Measurement","volume":"1","author":"Bland","year":"1986","journal-title":"Lancet"},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"378","DOI":"10.1037\/h0031619","article-title":"Measuring Nominal Scale Agreement among Many Raters","volume":"76","author":"Fleiss","year":"1971","journal-title":"Psychol. Bull."},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"289","DOI":"10.1111\/j.2517-6161.1995.tb02031.x","article-title":"Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing","volume":"57","author":"Benjamini","year":"1995","journal-title":"R. Stat. Soc. J. Ser. B Methodol."},{"key":"ref_43","unstructured":"Zhang, J., Elluri, S., Cherukuvada, B., Joffe, Y., Sena, J., Contreras, M., Siegel, S., Nerella, S., Price, C.E., and Rashidi, P. (2026). Auditing Multimodal LLM Raters: Central Tendency Bias in Clinical Ordinal Scoring. arXiv."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"42","DOI":"10.1097\/BCR.0B013E31802C88B2","article-title":"Effects of Differences in Percent Total Body Surface Area Estimation on Fluid Resuscitation of Transferred Burn Patients","volume":"28","author":"Freiburg","year":"2007","journal-title":"J. Burn Care Res."},{"key":"ref_45","unstructured":"Deutsche Gesellschaft f\u00fcr Verbrennungsmedizin (2026, August 02). Behandlung von Thermischen Verletzungen Bei Erwachsenen, Version 8.0 (Treatment of Thermal Injuries in Adults). Available online: https:\/\/register.awmf.org\/de\/leitlinien\/detail\/044-001."},{"key":"ref_46","unstructured":"Deutsche Gesellschaft f\u00fcr Kinderchirurgie (2026, August 02). Behandlung Thermischer Verletzungen Im Kindesalter (Verbrennung, Verbr\u00fchung), Version 3.0 (Treatment of Thermal Injuries in Childhood: Burns and Scalds). Available online: https:\/\/register.awmf.org\/de\/leitlinien\/detail\/006-128."},{"key":"ref_47","doi-asserted-by":"crossref","first-page":"874","DOI":"10.1111\/j.1749-6632.1968.tb14738.x","article-title":"Physiological Response to Crystalloid Resuscitation of Severe Burns","volume":"150","author":"Baxter","year":"1968","journal-title":"Ann. N. Y. Acad. Sci."},{"key":"ref_48","unstructured":"Stanford Institute for Human-Centered Artificial Intelligence (2026, August 02). The 2026 AI Index Report. Available online: https:\/\/hai.stanford.edu\/ai-index\/2026-ai-index-report."},{"key":"ref_49","unstructured":"Sutton, R.S. (2026, August 02). The Bitter Lesson. Available online: http:\/\/www.incompleteideas.net\/IncIdeas\/BitterLesson.html."},{"key":"ref_50","unstructured":"Reed, S., Zolna, K., Parisotto, E., Colmenarejo, S.G., Novikov, A., Barth-Maron, G., Gimenez, M., Sulsky, Y., Kay, J., and Springenberg, J.T. (2022). A Generalist Agent. Trans. Mach. Learn. Res."},{"key":"ref_51","unstructured":"Peled-Cohen, L., Calderon, N., Lissak, S., and Reichart, R. (2024). Learning the Bitter Lesson: Empirical Evidence from 20 Years of CVPR Proceedings. Proceedings of the 1st Workshop on NLP for Science (NLP4Science), Association for Computational Linguistics."}],"container-title":["Bioengineering"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2306-5354\/13\/9\/1000\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,28]],"date-time":"2026-08-28T13:02:25Z","timestamp":1787922145000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2306-5354\/13\/9\/1000"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,28]]},"references-count":51,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2026,9]]}},"alternative-id":["bioengineering13091000"],"URL":"https:\/\/doi.org\/10.3390\/bioengineering13091000","relation":{},"ISSN":["2306-5354"],"issn-type":[{"value":"2306-5354","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,28]]}}}