[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-379299-105":3,"detail-sidebar-cat-0-en-105":85,"doc-detail-379299-en":134},{"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":78,"head_meta":80,"extra_data":82,"updated_unix":84},105,"en","an-explainable-artificial-intelligence-analysis-of-raman-spectra-for-thyroid-cancer-diagnosis","An eXplainable Artificial Intelligence analysis of Raman spectra for thyroid cancer diagnosis","","Raman spectroscopy provides a non-invasive, label/dye-free way to detect biochemical changes in thyroid cancer and to discriminate malignant from benign nodules. The study develops a machine-learning pipeline using data-driven, label-independent peak selection from Raman spectra of histological samples. Models achieve strong discrimination, with performance measured by AUC greater than 0.9. eXplainable Artificial Intelligence is used to compute feature contributions per sample, improving interpretability and supporting clinical diagnostic decision-making.",{"@graph":14,"@context":77},[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/healthcare/","Healthcare",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/an-explainable-artificial-intelligence-analysis-of-raman-spectra-for-thyroid-cancer-diagnosis/379299/",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/an-explainable-artificial-intelligence-analysis-of-raman-spectra-for-thyroid-cancer-diagnosis/379299.png","ImageObject",300,407,{"name":42,"@type":43},"Stanford","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-27","2026-09-24",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",9,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69,73],{"name":60,"@type":61,"acceptedAnswer":62},"What problem does the study address in thyroid cancer diagnosis?","Question",{"text":63,"@type":64},"The study targets the difficulty of identifying significant Raman spectral patterns and peaks to reliably distinguish healthy/benign from malignant thyroid nodules, while keeping results interpretable.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"How is the Raman spectroscopy data used in the proposed machine-learning approach?",{"text":68,"@type":64},"Raman spectra from histological samples are processed with a data-driven, label-independent method to select a set of peaks, and machine-learning models are trained using the relative prominence of those peaks.",{"name":70,"@type":61,"acceptedAnswer":71},"How accurate are the models and how is performance measured?",{"text":72,"@type":64},"Model discrimination is quantified using the area under the Receiver Operating Characteristic curve, and the reported performance exceeds 0.9.",{"name":74,"@type":61,"acceptedAnswer":75},"Why is eXplainable Artificial Intelligence included?",{"text":76,"@type":64},"eXplainable Artificial Intelligence is used to compute each feature’s contribution to predictions for each sample, enhancing interpretability of which spectral features drive the model output.","https://schema.org",{"og:url":32,"og:type":79,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":81,"canonical":32},"index,follow",{"doc_id":83,"site_id":7},379299,1790532560,{"code":4,"msg":86,"data":87},"success",[88,92,96,100,105,110,114,119,123,126,130],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":97,"show_sort_weight":98,"slug":99},"Exam",70,"exam",{"id":101,"doc_module":4,"doc_module_name":25,"category_name":102,"show_sort_weight":103,"slug":104},5,"Comic",60,"comic",{"id":106,"doc_module":4,"doc_module_name":25,"category_name":107,"show_sort_weight":108,"slug":109},6,"Technology",50,"technology",{"id":111,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":112,"slug":113},7,40,"healthcare",{"id":115,"doc_module":4,"doc_module_name":25,"category_name":116,"show_sort_weight":117,"slug":118},8,"Research & Report",30,"research-report",{"id":55,"doc_module":4,"doc_module_name":25,"category_name":120,"show_sort_weight":121,"slug":122},"Religion & Spirituality",20,"religion-spirituality",{"id":121,"doc_module":4,"doc_module_name":25,"category_name":124,"show_sort_weight":121,"slug":125},"World Cup","world-cup",{"id":127,"doc_module":4,"doc_module_name":25,"category_name":128,"show_sort_weight":127,"slug":129},10,"Lifestyle","lifestyle",{"id":131,"doc_module":4,"doc_module_name":25,"category_name":132,"show_sort_weight":101,"slug":133},19,"General","general",{"code":4,"msg":86,"data":135},{"doc_id":83,"user_id":136,"nickname":42,"user_avatar":137,"doc_module":4,"category_id":111,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":138,"file_id":139,"file_url":140,"file_type":141,"file_size":142,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":143,"language":144,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":145,"faqs":146,"seo_title":147,"seo_description":12,"update_tm":148,"read_time":149},2336477552062,"https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d","[www. nature.com/scientificreports](www. nature.com/scientificreports)  \nOPEN  \nAn eXplainable Artificial Intelligence analysis of Raman spectra for thyroid cancer diagnosis  \nLoredana Bellantuono1,2, Raffaele Tommasi1, Ester Pantaleo2,3, Martina Verri4,5, Nicola Amoroso2,6, Pierfilippo Crucitti7, Michael Di Gioacchino5*, Filippo Longo7, Alfonso Monaco2,3, Anda Mihaela Naciu8, Andrea Palermo8, Chiara Taffon4, Sabina Tangaro2,9, Anna Crescenzi4, Armida Sodo5 & Roberto Bellotti2,3  \nRaman spectroscopy shows great potential as a diagnostic tool for thyroid cancer due to its ability to detect biochemical changes during cancer development. This technique is particularly valuable because it is non-invasive and label/dye-free. Compared to molecular tests, Raman spectroscopy analyses can more effectively discriminate malignant features, thus reducing unnecessary surgeries. However, one major hurdle to using Raman spectroscopy as a diagnostic tool is the identification of significant patterns and peaks. In this study, we propose a Machine Learning procedure to discriminate healthy/benign versus malignant nodules that produces interpretable results. We collect Raman spectra obtained from histological samples, select a set of peaks with a data-driven and label independent approach and train the algorithms with the relative prominence of the peaks in the selected set. The performance of the considered models, quantified by area under the Receiver Operating Characteristic curve, exceeds 0.9. To enhance the interpretability of the results, we employ eXplainable Artificial Intelligence and compute the contribution of each feature to the prediction of each sample.  \nThyroid cancer, consisting in the malignant growth of cells within the thyroid gland, is the most common malignant neoplasia of the endocrine system. Incidence rates, varying worldwide but generally placing it among the ten most prevalent cancers, have increased during the past decades, mostly due to an improvement in diagnostic procedures. The three main types commonly observed of thyroid follicular epithelial cell-derived cancer are papillary thyroid carcinoma (PTC), follicular carcinoma (FC), and the follicular variant of papillary thyroid carcinoma (FV-PTC) . PTC is characterized by the presence of papillary structures; it predominantly affects younger individuals and is the most prevalent, accounting for approximately 􀀞􀀝of cases. FC is the second most common type, representing around 10–15 of cases; it primarily affects older individuals and is more prevalent in areas with iodine deficiency. FV-PTC shares many characteristics with PTC but exhibits a follicular growth pattern, which poses challenges in distinguishing it from FC. Given the significantly high 5-year relative survival rate in early stages, the importance of efficient diagnostic methods cannot be overstated1.  \nEmerging issues in clinical practice include the global increase in detection of thyroid nodules and the consequent rise in the diagnosis of small carcinomas. Other specific challenges are constituted by deciding the extent of surgical treatment and the management of cytologically indeterminate thyroid nodules, and by interobserver diagnosis variability2–7. In addition, during the histological assessment of surgically excised thyroid glands, tumors with a follicular pattern can present diagnostic issues, since the evidence of malignant features, such as capsular or vascular invasion, may not be sufficient; these cases can be classified as follicular tumors of uncertain malignant potential (FT-UMP)8, thus leading to a questionable evaluation of the patient’s risk. Over  \n1Dipartimento di Biomedicina Traslazionale e Neuroscienze (DiBraiN), Università degli Studi di Bari Aldo Moro, 70124 Bari, Italy. 2Istituto Nazionale di Fisica Nucleare, Sezione di Bari, 70125 Bari, Italy. 3Dipartimento Interateneo di Fisica, Università degli Studi di Bari Aldo Moro, 70125 Bari, Italy. 4Unit of Endocrine Organs and Neuromusc","cbCair3xcrd5H28s","https://ap.wps.com/l/cbCair3xcrd5H28s","pdf",2460413,15,"English","# Introduction\n## Raman spectroscopy for thyroid cancer diagnosis\n## Clinical challenges and tumor subtypes\n## Limits of molecular testing and need for new tools\n# Methods and Explainability\n## Peak selection and machine-learning training\n## Model performance and ROC/AUC evaluation\n## eXplainable Artificial Intelligence feature contributions","[{\"question\":\"What problem does the study address in thyroid cancer diagnosis?\",\"answer\":\"The study targets the difficulty of identifying significant Raman spectral patterns and peaks to reliably distinguish healthy/benign from malignant thyroid nodules, while keeping results interpretable.\"},{\"question\":\"How is the Raman spectroscopy data used in the proposed machine-learning approach?\",\"answer\":\"Raman spectra from histological samples are processed with a data-driven, label-independent method to select a set of peaks, and machine-learning models are trained using the relative prominence of those peaks.\"},{\"question\":\"How accurate are the models and how is performance measured?\",\"answer\":\"Model discrimination is quantified using the area under the Receiver Operating Characteristic curve, and the reported performance exceeds 0.9.\"},{\"question\":\"Why is eXplainable Artificial Intelligence included?\",\"answer\":\"eXplainable Artificial Intelligence is used to compute each feature’s contribution to predictions for each sample, enhancing interpretability of which spectral features drive the model output.\"}]","An eXplainable Artificial Intelligence analysis of Raman spectra for thyroid cancer diagnosis | PDF",1790234992,38]