[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-333540-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-333540-en":130},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","deep-learning-based-cell-segmentation-for-rapid-optical-cytopathology-of-thyroid-cancer","Deep learning-based cell segmentation for rapid optical cytopathology of thyroid cancer","","Deep learning-based automated cell segmentation accelerates quantitative fluorescence polarization (Fpol) imaging using methylene blue (MB) for thyroid cancer detection. A 2DU-Net convolutional neural network was trained and tested on pathologically diverse human thyroid cell images. Performance was evaluated by comparing selected cell counts, segmented areas, and resulting Fpol values generated with automated (AU) versus manual (MA) data processing. The model segmented 15.8% more cells, achieved an auto-processing time of 10 s versus one hour for MA processing, and showed no statistically significant differences in AU- and MA-derived Fpol data.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/deep-learning-based-cell-segmentation-for-rapid-optical-cytopathology-of-thyroid-cancer/333540/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/deep-learning-based-cell-segmentation-for-rapid-optical-cytopathology-of-thyroid-cancer/333540.png","ImageObject",300,407,{"name":42,"@type":43},"Maeve","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-29","2026-09-21",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",5,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"What problem does the method address in thyroid cancer diagnosis?","Question",{"text":63,"@type":64},"It addresses the need to reduce data analysis time for MB fluorescence polarization imaging by automating cell segmentation using deep learning.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"How was the segmentation model evaluated against human processing?",{"text":68,"@type":64},"The evaluation compared cell counts, segmented areas, and fluorescence polarization (Fpol) values derived from automated (AU) versus manual (MA) data processing.",{"name":70,"@type":61,"acceptedAnswer":71},"What were the main performance findings for the 2DU-Net model?",{"text":72,"@type":64},"The model segmented 15.8% more cells than the human operator, reduced auto-processing time to 10 seconds versus one hour for manual processing, and showed no statistically significant differences in AU- versus MA-derived Fpol data.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},333540,1791387735,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,100,105,110,114,119,122,126],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":55,"doc_module":4,"doc_module_name":25,"category_name":97,"show_sort_weight":98,"slug":99},"Comic",60,"comic",{"id":101,"doc_module":4,"doc_module_name":25,"category_name":102,"show_sort_weight":103,"slug":104},6,"Technology",50,"technology",{"id":106,"doc_module":4,"doc_module_name":25,"category_name":107,"show_sort_weight":108,"slug":109},7,"Healthcare",40,"healthcare",{"id":111,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":112,"slug":113},8,30,"research-report",{"id":115,"doc_module":4,"doc_module_name":25,"category_name":116,"show_sort_weight":117,"slug":118},9,"Religion & Spirituality",20,"religion-spirituality",{"id":117,"doc_module":4,"doc_module_name":25,"category_name":120,"show_sort_weight":117,"slug":121},"World Cup","world-cup",{"id":123,"doc_module":4,"doc_module_name":25,"category_name":124,"show_sort_weight":123,"slug":125},10,"Lifestyle","lifestyle",{"id":127,"doc_module":4,"doc_module_name":25,"category_name":128,"show_sort_weight":55,"slug":129},19,"General","general",{"code":4,"msg":82,"data":131},{"doc_id":79,"user_id":132,"nickname":42,"user_avatar":133,"doc_module":4,"category_id":111,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":139,"language":140,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":12,"update_tm":144,"read_time":145,"preview_page_count":4},5909877438554,"https://ap-avatar.wpscdn.com/avatar/5600025385ad2bf12a7?_k=1778553567797529272","[www. nature.com/scientificreports](www. nature.com/scientificreports)  \nOPEN  \nDeep learning‑based cell segmentation for rapid optical cytopathology of thyroid cancer  \nPeter R. Jermain1,2, Martin Oswald3, Tenzin Langdun3, Santana Wright1, Ashraf Khan4, Thilo Stadelmann3,5, Ahmed Abdulkadir3,6 & Anna N. Yaroslavsky1,7*  \nFluorescence polarization (Fpol) imaging of methylene blue (MB) is a promising quantitative approach to thyroid cancer detection. Clinical translation of MB Fpol technology requires reduction of the data analysis time that can be achieved via deep learning‑based automated cell segmentation with a 2DU‑Net convolutional neural network. The model was trained and tested using images of pathologically diverse human thyroid cells and evaluated by comparing the number of cells selected, segmented areas, and Fpol values obtained using automated (AU) and manual (MA) data processing methods. Overall, the model segmented 15.8% more cells than the human operator. Differences in AU and MA segmented cell areas varied between − 55.2 and + 31.0%, whereas differences in Fpol values varied  \nfrom − 20.7 and + 10.7%. No statistically significant differences between AU and MA derived Fpol data were observed. The largest differences in Fpol values correlated with greatest discrepancies in AU versus MA segmented cell areas. Time required for auto‑processing was reduced to 10 s versus one hour required for MA data processing. Implementation of the automated cell analysis makes quantitative fluorescence polarization‑based diagnosis clinically feasible.  \nKeywords Thyroid cancer, Cytopathology, Fluorescence polarization, Methylene blue, Automated cell segmentation, Semantic segmentation  \nThyroid cancer is the 9th most common malignancy worldwide1. Approximately 44,000 new cases are expected to be diagnosed this year in the United States1,2. Besides, studies show that over 60% of adults will develop one or more benign thyroid nodules during their lifetime3. Benign nodule growth can be influenced by various health factors (e.g., age, genetics, hormonal changes, etc.) and only about 1% will result in thyroid disease4. Prognosis for malignant thyroid nodules largely depends on the type of cancer, tumor size and stage at diagnosis, as well as subsequent medical intervention5,6. Therefore, early and accurate differentiation of cancerous versus noncancerous neoplasms is imperative for guiding clinical management of thyroid tumors7,8.  \nFine-needle aspiration (FNA) cytopathology is an established clinical method for detecting thyroid cancer9–14. According to the Bethesda System for Reporting Thyroid Cytopathology (TBSRTC), thyroid FNAs are divided into six categories: (I) non-diagnostic, (II) benign,(III)–(V) indeterminate, which include atypical, undetermined, and suspicious for malignancy specimens, and (VI) malignant15–19. Definitive assessment of cytologically indeterminate samples (TBSRTC III–V), which represent 40% of thyroid FNA cases, requires histological evaluation of surgical specimens following lobectomy or complete thyroidectomy20,21. Surgical complications such as laryngeal nerve paralysis (5–11% of patients) or hypocalcemia frequently occur (20–30%)21–23. In all cases where surgical resection is completed, patients must be subjected to hormone replacement therapy for the remainder of their lifetime23,24.  \nRecently, molecular genetic testing has been implemented for the cytologically indeterminate thyroid nodules25,26. These assays may be able to detect malignant thyroid cells, determine the subtype of thyroid cancer, and/or provide information concerning the disease prognosis by identifying specific chromosomal abnormalities or gene mutations27,28. While some reports have indicated that genomic profiling can boost diagnostic  \n1Advanced Biophotonics Laboratory, University of Massachusetts Lowell, Lowell, MA, USA. 2Department of Radiation Medicine, MedStar Georgetown University Hospital, Washington, DC, USA. 3Centre for Artif","cbCaijVmJq8zw4wQ","https://ap.wps.com/l/cbCaijVmJq8zw4wQ","pdf",1651892,11,"English","# Background and clinical need\n## Thyroid cancer burden and diagnostic challenge\n## Limits of fine-needle aspiration (FNA) and surgery-based confirmation\n# Existing and emerging diagnostic approaches\n## Molecular genetic testing for indeterminate nodules\n## Fluorescence polarization imaging and methylene blue (MB)\n# Proposed method\n## Deep learning-based automated cell segmentation (2DU-Net)\n## Training, testing, and evaluation strategy\n# Results and clinical implications\n## Segmentation accuracy vs human operators\n## Agreement of Fpol values and processing time reduction","[{\"question\":\"What problem does the method address in thyroid cancer diagnosis?\",\"answer\":\"It addresses the need to reduce data analysis time for MB fluorescence polarization imaging by automating cell segmentation using deep learning.\"},{\"question\":\"How was the segmentation model evaluated against human processing?\",\"answer\":\"The evaluation compared cell counts, segmented areas, and fluorescence polarization (Fpol) values derived from automated (AU) versus manual (MA) data processing.\"},{\"question\":\"What were the main performance findings for the 2DU-Net model?\",\"answer\":\"The model segmented 15.8% more cells than the human operator, reduced auto-processing time to 10 seconds versus one hour for manual processing, and showed no statistically significant differences in AU- versus MA-derived Fpol data.\"}]","Deep learning-based cell segmentation for rapid optical cytopathology of thyroid cancer | PDF",1790002907,28]