{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,3]],"date-time":"2026-08-03T19:08:40Z","timestamp":1785784120745,"version":"3.56.0"},"reference-count":34,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2024,11,6]],"date-time":"2024-11-06T00:00:00Z","timestamp":1730851200000},"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. Endocrinol."],"abstract":"<jats:sec><jats:title>Background<\/jats:title><jats:p>Informative biomarkers play a vital role in guiding clinical decisions regarding management of cancers. We have previously demonstrated the potential of a deep convolutional neural network (CNN) for predicting cancer driver gene mutations from expert-curated histopathologic images in papillary thyroid carcinomas (PTCs). Recognizing the importance of whole slide image (WSI) analysis for clinical application, we aimed to develop an automated image preprocessing workflow that uses WSI inputs to categorize PTCs based on driver mutations.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>Histopathology slides from The Cancer Genome Atlas (TCGA) repository were utilized for diagnostic purposes. These slides underwent an automated tile extraction and preprocessing pipeline to ensure analysis-ready quality. Next, the extracted image tiles were utilized to train a deep learning CNN model, specifically Google\u2019s Inception v3, for the classification of PTCs. The model was trained to distinguish between different groups based on <jats:italic>BRAF<jats:sup>V600E<\/jats:sup><\/jats:italic> or <jats:italic>RAS<\/jats:italic> mutations.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>The newly developed pipeline performed equally well as the expert-curated image classifier. The best model achieved Area Under the Curve (AUC) values of 0.86 (ranging from 0.847 to 0.872) for validation and 0.865 (ranging from 0.854 to 0.876) for the final testing subsets. Notably, it accurately predicted 90% of tumors in the validation set and 84.2% in the final testing set. Furthermore, the performance of our new classifier showed a strong correlation with the expert-curated classifier (Spearman rho = 0.726, p = 5.28 e-08), and correlated with the molecular expression-based classifier, BRS (BRAF-RAS scores) (Spearman rho = 0.418, p = 1.92e-13).<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusions<\/jats:title><jats:p>Utilizing WSIs, we implemented an automated workflow with deep CNN model that accurately classifies driver mutations in PTCs.<\/jats:p><\/jats:sec>","DOI":"10.3389\/fendo.2024.1395979","type":"journal-article","created":{"date-parts":[[2024,11,5]],"date-time":"2024-11-05T07:08:02Z","timestamp":1730790482000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":5,"title":["Classifying driver mutations of papillary thyroid carcinoma on whole slide image: an automated workflow applying deep convolutional neural network"],"prefix":"10.3389","volume":"15","author":[{"given":"Peiling","family":"Tsou","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chang-Jiun","family":"Wu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2024,11,6]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.1002\/cncr.24416","article-title":"Increasing incidence of differentiated thyroid cancer in the United States, 1988-2005","volume":"115","author":"Chen","year":"2009","journal-title":"Cancer"},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2014.09.050","article-title":"Integrated genomic characterization of papillary thyroid carcinoma","volume":"159","year":"2014","journal-title":"Cell"},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.1634\/theoncologist.2013-0072","article-title":"RAS mutations in thyroid cancer","volume":"18","author":"Howell","year":"2013","journal-title":"Oncologist"},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.1677\/erc.1.0978","article-title":"BRAF mutation in thyroid cancer","volume":"12","author":"Xing","year":"2005","journal-title":"Endocr Relat Cancer"},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.1210\/jc.2006-2707","article-title":"BRAF mutations in papillary thyroid carcinomas inhibit genes involved in iodine metabolism","volume":"92","author":"Durante","year":"2007","journal-title":"J Clin Endocrinol Metab"},{"key":"B6","doi-asserted-by":"publisher","DOI":"10.1001\/jama.2013.3190","article-title":"Association between BRAF V600E mutation and mortality in patients with papillary thyroid cancer","volume":"309","author":"Xing","year":"2013","journal-title":"JAMA"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.1210\/jc.2008-0607","article-title":"BRAF(V600E) mutation and outcome of patients with papillary thyroid carcinoma: a 15-year median follow-up study","volume":"93","author":"Elisei","year":"2008","journal-title":"J Clin Endocrinol Metab"},{"key":"B8","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.0900780106","article-title":"(V600E)BRAF is associated with disabled feedback inhibition of RAF-MEK signaling and elevated transcriptional output of the pathway","volume":"106","author":"Pratilas","year":"2009","journal-title":"Proc Natl Acad Sci U.S.A"},{"key":"B9","first-page":"1","article-title":"Nodule detection in infrared thermography using deep learning","author":"Pathak","year":"2022"},{"key":"B10","doi-asserted-by":"publisher","first-page":"9428","DOI":"10.1038\/s41598-022-13541-2","article-title":"Deep learning features encode interpretable morphologies within histological images","volume":"12","author":"Foroughi Pour","year":"2022","journal-title":"Sci Rep"},{"key":"B11","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-021-01620-2","article-title":"Artificial intelligence for diagnosis and Gleason grading of prostate cancer: the PANDA challenge","volume":"28","author":"Bulten","year":"2022","journal-title":"Nat Med"},{"key":"B12","doi-asserted-by":"publisher","DOI":"10.1038\/s43018-022-00436-4","article-title":"Artificial intelligence in histopathology: enhancing cancer research and clinical oncology","volume":"3","author":"Shmatko","year":"2022","journal-title":"Nat Cancer"},{"key":"B13","doi-asserted-by":"publisher","DOI":"10.1038\/s41571-019-0252-y","article-title":"Artificial intelligence in digital pathology - new tools for diagnosis and precision oncology","volume":"16","author":"Bera","year":"2019","journal-title":"Nat Rev Clin Oncol"},{"key":"B14","doi-asserted-by":"publisher","first-page":"17","DOI":"10.4103\/jpi.jpi_91_18","article-title":"And they said it couldn\u2019t be done: predicting known driver mutations from H&E slides","volume":"10","author":"Montalto","year":"2019","journal-title":"J Pathol Inform"},{"key":"B15","doi-asserted-by":"publisher","first-page":"1650","DOI":"10.1016\/j.ccell.2023.08.002","article-title":"Transformer-based biomarker prediction from colorectal cancer histology: A large-scale multicentric study","volume":"41","author":"Wagner","year":"2023","journal-title":"Cancer Cell"},{"key":"B16","doi-asserted-by":"publisher","DOI":"10.1038\/s43018-020-0085-8","article-title":"Pan-cancer computational histopathology reveals mutations, tumor composition and prognosis","volume":"1","author":"Fu","year":"2020","journal-title":"Nat Cancer"},{"key":"B17","doi-asserted-by":"publisher","DOI":"10.1007\/s10278-020-00351-z","article-title":"Whole slide imaging (WSI) in pathology: current perspectives and future directions","volume":"33","author":"Kumar","year":"2020","journal-title":"J Digit Imaging"},{"key":"B18","doi-asserted-by":"publisher","DOI":"10.1038\/s43018-020-0087-6","article-title":"Pan-cancer image-based detection of clinically actionable genetic alterations","volume":"1","author":"Kather","year":"2020","journal-title":"Nat Cancer"},{"key":"B19","doi-asserted-by":"publisher","first-page":"1675","DOI":"10.3390\/jcm8101675","article-title":"Mapping driver mutations to histopathological subtypes in papillary thyroid carcinoma: applying a deep convolutional neural network","volume":"8","author":"Tsou","year":"2019","journal-title":"J Clin Med"},{"key":"B20","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-018-0177-5","article-title":"Classification and mutation prediction from non-small cell lung cancer histopathology images using deep learning","volume":"24","author":"Coudray","year":"2018","journal-title":"Nat Med"},{"key":"B21","doi-asserted-by":"publisher","first-page":"353","DOI":"10.1186\/s12859-023-05474-y","article-title":"HAHNet: a convolutional neural network for HER2 status classification of breast cancer","volume":"24","author":"Wang","year":"2023","journal-title":"BMC Bioinf"},{"key":"B22","doi-asserted-by":"publisher","first-page":"96","DOI":"10.1097\/CMR.0000000000000421","article-title":"Rapid BRAF mutation tests in patients with advanced melanoma: comparison of immunohistochemistry, Droplet Digital PCR, and the Idylla Mutation Platform","volume":"28","author":"Bisschop","year":"2018","journal-title":"Melanoma Res"},{"key":"B23","doi-asserted-by":"publisher","first-page":"4404","DOI":"10.1038\/s41598-023-31284-6","article-title":"Predicting EGFR mutational status from pathology images using a real-world dataset","volume":"13","author":"Pao","year":"2023","journal-title":"Sci Rep"},{"key":"B24","doi-asserted-by":"publisher","first-page":"39","DOI":"10.1097\/PAS.0000000000000948","article-title":"Whole slide imaging versus microscopy for primary diagnosis in surgical pathology: A multicenter blinded randomized noninferiority study of 1992 cases (Pivotal study)","volume":"42","author":"Mukhopadhyay","year":"2018","journal-title":"Am J Surg Pathol"},{"key":"B25","unstructured":"FDA allows marketing of first whole slide imaging system for digital pathology2017"},{"key":"B26","doi-asserted-by":"publisher","DOI":"10.1002\/path.5898","article-title":"Artificial intelligence to identify genetic alterations in conventional histopathology","volume":"257","author":"Cifci","year":"2022","journal-title":"J Pathol"},{"key":"B27","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1200\/CCI.18.00157","article-title":"HistoQC: an open-source quality control tool for digital pathology slides","volume":"3","author":"Janowczyk","year":"2019","journal-title":"JCO Clin Cancer Inform"},{"key":"B28","unstructured":"AmgadM\n            BeezeleyJ\n            ChittajalluDR\n            CooperL\n            GutmanD\n            HelbaB\n          HistomicsTK2016"},{"key":"B29","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR.2016.308","article-title":"Rethinking the inception architecture for computer vision","author":"Szegedy","year":"2016"},{"key":"B30","doi-asserted-by":"publisher","DOI":"10.1016\/j.ebiom.2017.12.026","article-title":"Deep convolutional neural networks enable discrimination of heterogeneous digital pathology images","volume":"27","author":"Khosravi","year":"2018","journal-title":"EBioMedicine"},{"key":"B31","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2020.101813","article-title":"Deep neural network models for computational histopathology: A survey","volume":"67","author":"Srinidhi","year":"2021","journal-title":"Med Image Anal"},{"key":"B32","doi-asserted-by":"publisher","DOI":"10.1007\/s11684-020-0782-9","article-title":"Deep learning in digital pathology image analysis: a survey","volume":"14","author":"Deng","year":"2020","journal-title":"Front Med"},{"key":"B33","doi-asserted-by":"publisher","first-page":"87","DOI":"10.1038\/s41698-021-00225-9","article-title":"Genetic mutation and biological pathway prediction based on whole slide images in breast carcinoma using deep learning","volume":"5","author":"Qu","year":"2021","journal-title":"NPJ Precis Oncol"},{"key":"B34","first-page":"1","article-title":"Sentiment analysis using machine learning and deep learning","author":"Chandra","year":"2020"}],"container-title":["Frontiers in Endocrinology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fendo.2024.1395979\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,11,5]],"date-time":"2024-11-05T07:08:10Z","timestamp":1730790490000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fendo.2024.1395979\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,11,6]]},"references-count":34,"alternative-id":["10.3389\/fendo.2024.1395979"],"URL":"https:\/\/doi.org\/10.3389\/fendo.2024.1395979","relation":{},"ISSN":["1664-2392"],"issn-type":[{"value":"1664-2392","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,11,6]]},"article-number":"1395979"}}