{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,18]],"date-time":"2026-10-18T09:25:00Z","timestamp":1792315500187,"version":"4.4.0"},"reference-count":29,"publisher":"Public Library of Science (PLoS)","issue":"3","license":[{"start":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T00:00:00Z","timestamp":1773878400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosdigitalhealth.org"],"crossmark-restriction":false},"short-container-title":["PLOS Digit Health"],"abstract":"<jats:p>Multimodal large language models (LLMs) with image recognition capabilities have emerged as potential tools for medical image analysis, yet their performance in specialized domains like histology remains largely unexplored. The objective of this study was to systematically evaluate the performance of leading multimodal LLMs in histological image interpretation and assess their visual recognition capabilities. Four multimodal LLMs (GPT-4o, Claude Sonnet 4, Gemini 2.5 Flash, and Copilot) were evaluated using 144 histological images representing four tissue types (epithelial, connective, muscle, and nervous) at three magnification levels. Each image was assessed using three standardized questions: tissue identification, morphological features, and functional analysis. Three expert faculty members independently graded responses using a 4-point scale (1\u2009=\u2009Poor to 4\u2009=\u2009Excellent). Friedman tests, ICC, and post-hoc power analyses were performed with statistical significance set at p\u2009&lt;\u2009.05. A clear performance hierarchy emerged with Gemini demonstrating superior performance (mean score: 3.35\/4.00), significantly outperforming all other models. Copilot and GPT-4o tied for second place (both 2.76\/4.00), while Claude showed the lowest performance (2.55\/4.00). Performance varied across tissue types, with epithelial tissue showing the greatest inter-model variation. Inter-rater reliability was good across all models (ICC\u2009&gt;\u20090.85), confirming assessment consistency. Post-hoc power analysis validated statistical significance for primary comparisons but indicated insufficient power to distinguish between the three lower-performing models. Current multimodal LLMs exhibit significant limitations in visual recognition relative to text processing performance. The substantial cross-modal performance gaps reveal some constraints in visual processing architectures, though the underlying mechanisms require further investigation. These findings establish technical benchmarks for multimodal LLM development and highlight the need for specialized visual processing innovations in their imaging processes.<\/jats:p>","DOI":"10.1371\/journal.pdig.0001306","type":"journal-article","created":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T17:27:53Z","timestamp":1773941273000},"page":"e0001306","update-policy":"https:\/\/doi.org\/10.1371\/journal.pdig.corrections_policy","source":"Crossref","is-referenced-by-count":5,"title":["Visual recognition limitations in multimodal large language models: A comparative analysis of histological image interpretation"],"prefix":"10.1371","volume":"5","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-1159-4573","authenticated-orcid":true,"given":"Volodymyr","family":"Mavrych","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8935-0183","authenticated-orcid":true,"given":"Einas M.","family":"Yousef","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ahmed","family":"Yaqinuddin","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Aftab Ahmed","family":"Shaikh","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-9496-9754","authenticated-orcid":true,"given":"Olena","family":"Bolgova","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,3,19]]},"reference":[{"issue":"4","key":"pdig.0001306.ref001","doi-asserted-by":"crossref","first-page":"538","DOI":"10.3390\/genes12040538","article-title":"Next-generation digital histopathology of the tumor microenvironment","volume":"12","author":"F Mungenast","year":"2021","journal-title":"Genes (Basel)"},{"issue":"11","key":"pdig.0001306.ref002","doi-asserted-by":"crossref","first-page":"100255","DOI":"10.1016\/j.labinv.2023.100255","article-title":"Revolutionizing digital pathology with the power of generative artificial intelligence and foundation models","volume":"103","author":"A Waqas","year":"2023","journal-title":"Lab Invest"},{"key":"pdig.0001306.ref003","doi-asserted-by":"crossref","first-page":"200","DOI":"10.1016\/j.ejca.2021.07.012","article-title":"Gastrointestinal cancer classification and prognostication from histology using deep learning: systematic review","volume":"155","author":"S Kuntz","year":"2021","journal-title":"Eur J Cancer"},{"key":"pdig.0001306.ref004","doi-asserted-by":"crossref","first-page":"226","DOI":"10.1016\/j.semcancer.2020.08.006","article-title":"Deep computational pathology in breast cancer","volume":"72","author":"A Duggento","year":"2021","journal-title":"Semin Cancer Biol"},{"issue":"1","key":"pdig.0001306.ref005","doi-asserted-by":"crossref","first-page":"72","DOI":"10.1038\/s41698-024-00573-2","article-title":"Large language models and multimodal foundation models for precision oncology","volume":"8","author":"D Truhn","year":"2024","journal-title":"NPJ Precis Oncol"},{"issue":"1","key":"pdig.0001306.ref006","doi-asserted-by":"crossref","DOI":"10.1148\/radiol.240609","article-title":"Evaluation of GPT large language model performance on RSNA 2023 case of the day questions","volume":"313","author":"P Mukherjee","year":"2024","journal-title":"Radiology"},{"key":"pdig.0001306.ref007","doi-asserted-by":"crossref","DOI":"10.2196\/57592","article-title":"Evaluating Bard Gemini Pro and GPT-4 vision against student performance in medical Visual Question Answering: comparative case study","volume":"8","author":"J Roos","year":"2024","journal-title":"JMIR Form Res"},{"key":"pdig.0001306.ref008","doi-asserted-by":"crossref","DOI":"10.2196\/55508","article-title":"Assessing the utility of multimodal large language models (GPT-4 vision and large language and vision assistant) in identifying melanoma across different skin tones","volume":"7","author":"K Cirone","year":"2024","journal-title":"JMIR Dermatol"},{"issue":"3","key":"pdig.0001306.ref009","doi-asserted-by":"crossref","first-page":"100681","DOI":"10.1016\/j.xops.2024.100681","article-title":"Large language models in ophthalmology: a review of publications from top ophthalmology journals","volume":"5","author":"AP Agnihotri","year":"2024","journal-title":"Ophthalmol Sci"},{"issue":"1","key":"pdig.0001306.ref010","doi-asserted-by":"crossref","first-page":"11","DOI":"10.1016\/j.diii.2022.11.007","article-title":"Artificial intelligence: a critical review of applications for lung nodule and lung cancer","volume":"104","author":"C de Margerie-Mellon","year":"2023","journal-title":"Diagn Interv Imaging"},{"key":"pdig.0001306.ref011","doi-asserted-by":"crossref","DOI":"10.2196\/59505","article-title":"Multimodal large language models in health care: applications, challenges, and future outlook","volume":"26","author":"R AlSaad","year":"2024","journal-title":"J Med Internet Res"},{"key":"pdig.0001306.ref012","doi-asserted-by":"crossref","DOI":"10.2196\/67244","article-title":"Large language models in biochemistry education: comparative evaluation of performance","volume":"11","author":"O Bolgova","year":"2025","journal-title":"JMIR Med Educ"},{"issue":"2","key":"pdig.0001306.ref013","doi-asserted-by":"crossref","first-page":"430","DOI":"10.1152\/advan.00093.2024","article-title":"Claude, ChatGPT, Copilot, and Gemini performance versus students in different topics of neuroscience","volume":"49","author":"V Mavrych","year":"2025","journal-title":"Adv Physiol Educ"},{"issue":"7","key":"pdig.0001306.ref014","doi-asserted-by":"crossref","first-page":"718","DOI":"10.1002\/ase.70044","article-title":"Comparative analysis of LLMs performance in medical embryology: a cross-platform study of ChatGPT, Claude, Gemini, and Copilot","volume":"18","author":"O Bolgova","year":"2025","journal-title":"Anat Sci Educ"},{"issue":"2","key":"pdig.0001306.ref015","doi-asserted-by":"crossref","first-page":"200","DOI":"10.1002\/ca.24244","article-title":"Using large language models (ChatGPT, Copilot, PaLM, Bard, and Gemini) in gross anatomy course: comparative analysis","volume":"38","author":"V Mavrych","year":"2025","journal-title":"Clin Anat"},{"key":"pdig.0001306.ref016","doi-asserted-by":"crossref","DOI":"10.2196\/45312","article-title":"How does ChatGPT perform on the United States Medical Licensing Examination (USMLE)? The implications of large language models for medical education and knowledge assessment","volume":"9","author":"A Gilson","year":"2023","journal-title":"JMIR Med Educ"},{"issue":"1","key":"pdig.0001306.ref017","doi-asserted-by":"crossref","first-page":"431","DOI":"10.1186\/s12909-024-05365-7","article-title":"Artificial intelligence and medical education: application in classroom instruction and student assessment using a pharmacology & therapeutics case study","volume":"24","author":"K Sridharan","year":"2024","journal-title":"BMC Med Educ"},{"key":"pdig.0001306.ref018","article-title":"Multimodal generative AI for anatomic pathology-a review of current applications to envisage the future direction","author":"E Ullah","year":"2025","journal-title":"Adv Anat Pathol"},{"issue":"2","key":"pdig.0001306.ref019","first-page":"287","article-title":"Evaluating ChatGPT-4\u2019s performance in identifying radiological anatomy in FRCR Part 1 examination questions","volume":"35","author":"PK Sarangi","year":"2024","journal-title":"Indian J Radiol Imaging"},{"issue":"5","key":"pdig.0001306.ref020","doi-asserted-by":"crossref","first-page":"507","DOI":"10.1177\/12034754241266166","article-title":"The performance of artificial intelligence Chatbot (GPT-4) on image-based dermatology certification board exam questions","volume":"28","author":"L Samman","year":"2024","journal-title":"J Cutan Med Surg"},{"issue":"11","key":"pdig.0001306.ref021","doi-asserted-by":"crossref","first-page":"1645","DOI":"10.1016\/j.jsurg.2024.08.002","article-title":"Artificial intelligence in orthopaedics: performance of ChatGPT on text and image questions on a complete AAOS Orthopaedic In-Training Examination (OITE)","volume":"81","author":"DS Hayes","year":"2024","journal-title":"J Surg Educ"},{"key":"pdig.0001306.ref022","doi-asserted-by":"crossref","DOI":"10.2196\/54793","article-title":"Curriculum frameworks and educational programs in AI for medical students, residents, and practicing physicians: scoping review","volume":"10","author":"R Tolentino","year":"2024","journal-title":"JMIR Med Educ"},{"issue":"3","key":"pdig.0001306.ref023","doi-asserted-by":"crossref","first-page":"372","DOI":"10.1002\/ase.2235","article-title":"Histology as a paradigm for a science-based learning experience: visits by histology education spirits of past, present, and future","volume":"16","author":"M Hortsch","year":"2023","journal-title":"Anat Sci Educ"},{"issue":"11","key":"pdig.0001306.ref024","doi-asserted-by":"crossref","first-page":"1276","DOI":"10.1111\/medu.15402","article-title":"A systematic review of large language models and their implications in medical education","volume":"58","author":"HC Lucas","year":"2024","journal-title":"Med Educ"},{"issue":"1","key":"pdig.0001306.ref025","doi-asserted-by":"crossref","first-page":"2534065","DOI":"10.1080\/10872981.2025.2534065","article-title":"Large language models in medical education: a comparative cross-platform evaluation in answering histological questions","volume":"30","author":"V Mavrych","year":"2025","journal-title":"Med Educ Online"},{"issue":"24","key":"pdig.0001306.ref026","doi-asserted-by":"crossref","first-page":"5040","DOI":"10.3390\/electronics13245040","article-title":"A review of large language models: fundamental architectures, key technological evolutions, interdisciplinary technologies integration, optimization and compression techniques, applications, and challenges","volume":"13","author":"S Han","year":"2024","journal-title":"Electronics"},{"issue":"9","key":"pdig.0001306.ref027","doi-asserted-by":"crossref","first-page":"1526","DOI":"10.1038\/s41562-023-01659-w","article-title":"Emergent analogical reasoning in large language models","volume":"7","author":"T Webb","year":"2023","journal-title":"Nat Hum Behav"},{"key":"pdig.0001306.ref028","doi-asserted-by":"crossref","first-page":"117370","DOI":"10.1016\/j.measurement.2025.117370","article-title":"Review on applications of computer vision techniques for pipeline inspection","volume":"252","author":"Y Liu","year":"2025","journal-title":"Measurement"},{"issue":"5","key":"pdig.0001306.ref029","doi-asserted-by":"crossref","first-page":"820","DOI":"10.1109\/JPROC.2021.3054390","article-title":"A review of deep learning in medical imaging: imaging traits, technology trends, case studies with progress highlights, and future promises","volume":"109","author":"SK Zhou","year":"2021","journal-title":"Proc IEEE Inst Electr Electron Eng"}],"container-title":["PLOS Digital Health"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0001306","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T17:27:58Z","timestamp":1773941278000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0001306"}},"subtitle":[],"editor":[{"given":"Jing","family":"Tian","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,3,19]]},"references-count":29,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2026,3,19]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pdig.0001306","relation":{},"ISSN":["2767-3170"],"issn-type":[{"value":"2767-3170","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,19]]}}}