{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,23]],"date-time":"2026-08-23T15:22:46Z","timestamp":1787498566156,"version":"build-2736575974"},"reference-count":32,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR AI"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n The visual similarity of melanoma and seborrheic keratosis has made it difficult for older patients with disabilities to know when to seek medical attention, contributing to the metastasis of melanoma.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to present a novel multimodal deep learning-based technique to distinguish between melanoma and seborrheic keratosis.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n Our strategy is three-fold: (1) use patient image data to train and test three deep learning models using transfer learning (ResNet50, InceptionV3, and VGG16) and one author-designed model, (2) use patient metadata to train and test a deep learning model, and (3) combine the predictions of the image model with the best accuracy and the metadata model, using nonlinear least squares regression to specify ideal weights to each model for a combined prediction.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n The accuracy of the combined model was 88% (195\/221 classified correctly) on test data from the HAM10000 dataset. Model reliability was assessed by visualizing the output activation map of each model and comparing the diagnosis patterns to that of dermatologists. The addition of metadata to the image dataset was key to reducing the false-negative and false-positive rates simultaneously, thereby producing better metrics and improving overall model accuracy.<\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n Results from this experiment could be used to eliminate late diagnosis of melanoma via easy access to an app. Future experiments can use text data (subjective data pertaining to how the patient felt over a certain period of time) to allow this model to reflect the real hospital setting to a greater extent.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/66561","type":"journal-article","created":{"date-parts":[[2025,7,6]],"date-time":"2025-07-06T21:05:08Z","timestamp":1751835908000},"page":"e66561-e66561","source":"Crossref","is-referenced-by-count":4,"title":["Deep Learning Multi-Modal Melanoma Detection: Algorithm Development and Validation"],"prefix":"10.2196","volume":"4","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-7927-1584","authenticated-orcid":false,"given":"Nithika","family":"Vivek","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2938-7942","authenticated-orcid":false,"given":"Karthik","family":"Ramesh","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2025,8,13]]},"reference":[{"issue":"10","key":"R1","doi-asserted-by":"publisher","DOI":"10.7759\/cureus.70697","article-title":"Trends in Melanoma incidence, prevalence, stage at diagnosis, and survival: an analysis of the United States Cancer Statistics (USCS) Database","volume":"16","author":"Okobi","journal-title":"Cureus"},{"key":"R2","doi-asserted-by":"publisher","DOI":"10.3389\/fmed.2023.1268479","article-title":"Advances in melanoma: epidemiology, diagnosis, and prognosis","volume":"10","author":"Waseh","journal-title":"Front Med (Lausanne)"},{"key":"R3","doi-asserted-by":"publisher","DOI":"10.2147\/TCRM.S316988","article-title":"Clinical and histopathological characteristics of tumors arising in seborrheic keratosis: a study of 1365 cases","volume":"17","author":"Ye","journal-title":"Ther Clin Risk Manag"},{"key":"R4","doi-asserted-by":"publisher","DOI":"10.12688\/f1000research.18983.1","article-title":"Recent advances in managing and understanding seborrheic keratosis","volume":"8","author":"Wollina","journal-title":"F1000Res"},{"issue":"2","key":"R5","doi-asserted-by":"publisher","first-page":"152","DOI":"10.5021\/ad.2016.28.2.152","article-title":"Clinical and histopathological investigation of seborrheic keratosis","volume":"28","author":"Roh","journal-title":"Ann Dermatol"},{"key":"R6","doi-asserted-by":"publisher","DOI":"10.2147\/CCID.S267246","article-title":"Differential diagnosis and management on seborrheic keratosis in elderly patients","volume":"14","author":"Moscarella","journal-title":"Clin Cosmet Investig Dermatol"},{"key":"R7","unstructured":"Heistein JB Acharya U Mukkamalla SKR . Malignant Melanoma. StatPearls Publishing; 2024. URL: https:\/\/www.ncbi.nlm.nih.gov\/books\/NBK470409 [Accessed 08-08-2025]"},{"issue":"2","key":"R8","doi-asserted-by":"publisher","first-page":"250","DOI":"10.1067\/mjd.2001.114598","article-title":"Dermatology in primary care: Prevalence and patient disposition","volume":"45","author":"Lowell","journal-title":"J Am Acad Dermatol"},{"issue":"Suppl 1","key":"R9","doi-asserted-by":"publisher","first-page":"15","DOI":"10.1038\/s41379-019-0402-x","article-title":"Melanoma pathology reporting and staging","volume":"33","author":"Scolyer","journal-title":"Mod Pathol"},{"issue":"11","key":"R10","doi-asserted-by":"publisher","first-page":"1366","DOI":"10.1080\/15384047.2019.1640032","article-title":"Current state of melanoma diagnosis and treatment","volume":"20","author":"Davis","journal-title":"Cancer Biol Ther"},{"issue":"13","key":"R11","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics13132139","article-title":"Accuracy of a smartphone-based artificial intelligence application for classification of melanomas, melanocytic nevi, and seborrheic keratoses","volume":"13","author":"Liutkus","journal-title":"Diagnostics (Basel)"},{"key":"R12","doi-asserted-by":"publisher","DOI":"10.3389\/fmed.2024.1331895","article-title":"Artificial intelligence and skin cancer","volume":"11","author":"Wei","journal-title":"Front Med (Lausanne)"},{"issue":"4","key":"R13","doi-asserted-by":"publisher","first-page":"348","DOI":"10.1034\/j.1600-0846.2003.00044.x","article-title":"Melanoma and seborrheic keratosis differentiation using texture features","volume":"9","author":"Deshabhoina","journal-title":"Skin Res Technol"},{"issue":"1","key":"R14","doi-asserted-by":"publisher","DOI":"10.1016\/j.xjidi.2022.100150","article-title":"Deep learning in dermatology: a systematic review of current approaches, outcomes, and limitations","volume":"3","author":"Jeong","journal-title":"JID Innov"},{"issue":"1","key":"R15","doi-asserted-by":"publisher","DOI":"10.1186\/s12880-020-00534-8","article-title":"Melanoma diagnosis using deep learning techniques on dermatoscopic images","volume":"21","author":"Jojoa Acosta","journal-title":"BMC Med Imaging"},{"issue":"8045","key":"R16","doi-asserted-by":"publisher","first-page":"319","DOI":"10.1038\/s41586-024-08328-6","article-title":"Accurate predictions on small data with a tabular foundation model","volume":"637","author":"Hollmann","journal-title":"Nature New Biol"},{"issue":"1","key":"R17","doi-asserted-by":"publisher","first-page":"109","DOI":"10.21015\/vtse.v11i1.1204","article-title":"Classification of skin disease using machine learning","volume":"11","author":"Ahmed","journal-title":"VFAST trans softw eng"},{"key":"R18","doi-asserted-by":"publisher","DOI":"10.22489\/CinC.2022.355","article-title":"Dual bayesian resnet: a deep learning approach to heart murmur detection","volume":"49","author":"Krones","journal-title":"Comput Cardiol Conf"},{"key":"R19","doi-asserted-by":"crossref","unstructured":"Wang Z Gao C Xiao C Sun JM . MediTab: scaling medical tabular data predictors via data consolidation, enrichment, and refinement. Presented at: Thirty-Third International Joint Conference on Artificial Intelligence {IJCAI-24}; Aug 3-9, 2024; Jeju, South Korea. URL: https:\/\/www.ijcai.org\/proceedings\/2024 [doi: 10.24963\/ijcai.2024\/670]","DOI":"10.24963\/ijcai.2024\/670"},{"key":"R20","doi-asserted-by":"crossref","unstructured":"Khademi S Hajiakhondi-Meybodi Z Vaseghi G Sarrafzadegan N Mohammadi A . FH-tabnet: multi-class familial hypercholesterolemia detection via a multi-stage tabular deep learning network. Presented at: 2024 32nd European Signal Processing Conference (EUSIPCO); Aug 26-30, 2024; Lyon, France. URL: https:\/\/eurasip.org\/Proceedings\/Eusipco\/Eusipco2024\/pdfs\/0001416.pdf [doi: 10.23919\/EUSIPCO63174.2024.10715254]","DOI":"10.23919\/EUSIPCO63174.2024.10715254"},{"issue":"2","key":"R21","doi-asserted-by":"publisher","first-page":"147","DOI":"10.1080\/00401706.2019.1708463","article-title":"Multiple tensor-on-tensor regression: an approach for modeling processes with heterogeneous sources of data","volume":"63","author":"Gahrooei","journal-title":"Technometrics"},{"key":"R22","doi-asserted-by":"crossref","unstructured":"Aoki Y Hayami K Toshimoto K Sugiyama Y . Cluster Gauss-Newton method for finding multiple approximate minimisers of nonlinear least squares problems with applications to parameter estimation of pharmacokinetic models. National Institute of Informatics; 2020. URL: https:\/\/www.nii.ac.jp\/TechReports\/public_html\/20-002E.pdf [Accessed 12-04-2024]","DOI":"10.1007\/s11081-020-09571-2"},{"issue":"5","key":"R23","doi-asserted-by":"publisher","first-page":"211","DOI":"10.4172\/2155-6180.1000211","article-title":"A New Robust Method for Nonlinear Regression","volume":"5","author":"Tabatabai","journal-title":"J Biom Biostat"},{"issue":"supplement 2","key":"R24","first-page":"241042","volume":"65","author":"Xia","journal-title":"J Nucl Med"},{"issue":"7639","key":"R25","doi-asserted-by":"publisher","first-page":"115","DOI":"10.1038\/nature21056","article-title":"Dermatologist-level classification of skin cancer with deep neural networks","volume":"542","author":"Esteva","journal-title":"Nature New Biol"},{"issue":"1","key":"R26","doi-asserted-by":"publisher","DOI":"10.1038\/sdata.2018.161","article-title":"The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions","volume":"5","author":"Tschandl","journal-title":"Sci Data"},{"issue":"12","key":"R27","doi-asserted-by":"publisher","first-page":"15923","DOI":"10.1007\/s10489-022-04278-6","article-title":"Front-end deep learning web apps development and deployment: a review","volume":"53","author":"Goh","journal-title":"Appl Intell"},{"key":"R28","doi-asserted-by":"publisher","DOI":"10.1155\/2017\/3762651","article-title":"Defining an optimal cut-point value in roc analysis: an alternative approach","volume":"2017","author":"Unal","journal-title":"Comput Math Methods Med"},{"issue":"1","key":"R29","doi-asserted-by":"publisher","first-page":"25","DOI":"10.4097\/kja.21209","article-title":"Receiver operating characteristic curve: overview and practical use for clinicians","volume":"75","author":"Nahm","journal-title":"Korean J Anesthesiol"},{"key":"R30","doi-asserted-by":"publisher","DOI":"10.3389\/frobt.2024.1444763","article-title":"Enhancing interpretability and accuracy of AI models in healthcare: a comprehensive review on challenges and future directions","volume":"11","author":"Ennab","journal-title":"Front Robot AI"},{"key":"R31","doi-asserted-by":"publisher","DOI":"10.1016\/j.dajour.2023.100230","article-title":"A systematic review of explainable artificial intelligence models and applications: recent developments and future trends","volume":"7","author":"A.","journal-title":"Decision Analytics Journal"},{"issue":"6","key":"R32","doi-asserted-by":"publisher","first-page":"771","DOI":"10.1684\/ejd.2021.4171","article-title":"Clinical ABCDE rule for early melanoma detection","volume":"31","author":"Duarte","journal-title":"Eur J Dermatol"}],"container-title":["JMIR AI"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/ai.jmir.org\/2025\/1\/e66561","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T22:28:40Z","timestamp":1785536920000},"score":1,"resource":{"primary":{"URL":"https:\/\/ai.jmir.org\/2025\/1\/e66561"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,8,13]]},"references-count":32,"URL":"https:\/\/doi.org\/10.2196\/66561","relation":{},"ISSN":["2817-1705"],"issn-type":[{"value":"2817-1705","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,8,13]]},"article-number":"v4i10e66561"}}