[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-352234-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-352234-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","dual-deep-learning-approach-for-noninvasive-renal-tumour-subtyping-with-verdict-mri","Dual deep learning approach for noninvasive renal tumour subtyping with VERDICT-MRI","","Renal cell carcinomas contain multiple subtypes that are difficult to distinguish using imaging alone. This study characterises renal tumour microstructure using diffusion MRI within the VERDICT-MRI framework, fitted with a self-supervised neural network in a three-compartment model. Fourteen prospective patients (17 renal tumours) were scanned with nine b-values, and VERDICT more accurately captured diffusion in tumour and healthy tissue. Feature selection identified a reduced four-b-value protocol, cutting scan time by over 30 minutes while enabling efficient imaging for larger cohorts.",{"@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/dual-deep-learning-approach-for-noninvasive-renal-tumour-subtyping-with-verdict-mri/352234/",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/dual-deep-learning-approach-for-noninvasive-renal-tumour-subtyping-with-verdict-mri/352234.png","ImageObject",300,407,{"name":42,"@type":43},"Logic","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-23","2026-09-22",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",10,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"What problem does this study address in renal tumour imaging?","Question",{"text":63,"@type":64},"Renal tumour subtypes are hard to distinguish using imaging alone, limiting accurate noninvasive characterization and risk assessment.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"How does VERDICT-MRI differ from simpler diffusion models?",{"text":68,"@type":64},"VERDICT uses a three-compartment biophysical model that better captures diffusion behaviour in both tumour and healthy tissue compared with ADC-based or simplified approaches.",{"name":70,"@type":61,"acceptedAnswer":71},"What scan-time reduction was achieved and how?",{"text":72,"@type":64},"A feature selection method produced a reduced 4 b-value protocol (b = [70, 150, 1000, 2000]), reducing scan time by over 30 minutes to support larger cohort studies.","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},352234,1790865089,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,111,115,120,123,126],{"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":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":108,"show_sort_weight":109,"slug":110},7,"Healthcare",40,"healthcare",{"id":112,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":113,"slug":114},8,30,"research-report",{"id":116,"doc_module":4,"doc_module_name":25,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":25,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":55,"doc_module":4,"doc_module_name":25,"category_name":124,"show_sort_weight":55,"slug":125},"Lifestyle","lifestyle",{"id":127,"doc_module":4,"doc_module_name":25,"category_name":128,"show_sort_weight":97,"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":112,"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":116,"language":139,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":140,"faqs":141,"seo_title":142,"seo_description":12,"update_tm":143,"read_time":144,"preview_page_count":33,"preview_urls":145},1099513958762,"https://ap-avatar.wpscdn.com/avatar/1000023916a998db790?x-image-process=image/resize,m_fixed,w_180,h_180&k=1784791008015729253","npj | imaging Article  \n\n| \u003Cbr>[https://doi.org/10.1038/s44303-025-00135-6](https://doi.org/10.1038/s44303-025-00135-6) |  |\n| --- | --- |\n| Dual deep learning approach for noninvasive renal tumour subtyping with VERDICT-MRI\u003Cbr> Check for updates |  |\n| Snigdha Sen1 , Lorna Smith2, Lucy Caselton3, Joey Clemente2, Maxine Tran4, Shonit Punwani2, David Atkinson2, Richard L. Hesketh5,6 & Eleftheria Panagiotaki1,6 |  |\n| Renal cell carcinomas (RCCs) have multiple subtypes that are difﬁcult to distinguish using imaging alone. This study characterises renal tumour microstructure using diffusion MRI (dMRI) and the Vascular, Extracellular and Restricted Diffusion for Cytometry in Tumours (VERDICT)-MRI framework. Patients were prospectively recruited from the RIM trial ([ClinicalTrials.gov: NCT07173140](ClinicalTrials.gov: NCT07173140) , 20/11/ 2024) . Fourteen patients with 17 renal tumours (including benign and various RCC subtypes) underwent dMRI using nine b-values (0–2500 s/mm²) . A three-compartment VERDICT model was ﬁtted with a self-supervised neural network. Compared to simpler dMRI models, VERDICT more accurately captured the diffusion data in tumour and healthy tissue. VERDICT revealed signiﬁcant differences in intracellular volume fraction between cancerous and normal tissue, and in vascular volume fraction between vascular and non-vascular regions. A feature selection method identiﬁed a reduced 4 b-value protocol (b = [70, 150, 1000, 2000]), cutting scan time by over 30 min, enabling more efﬁcient imaging in larger cohorts. |  |\n| Renal cancer accounts for roughly 2% of all cancer diagnoses and deaths worldwide1. Over 90% of primary renal cancers are renal cell carcinoma (RCC), arising from the renal tubule epithelium2. The 2022 WHO classiﬁcation subdividesRCC into21different histotypes ofwhich the commonest are clear-cell RCC (ccRCC) (~70%), papillary RCC (pRCC) (~10%) and chromophobe (chRCC) (~5%)3. Benign tumours of the kidney can also occur, notably oncocytoma and angiomyolipoma4. The main diagnostic imaging for renal tumours is computed tomography (CT), which can diagnose RCC with a sensitivity of over 95% and a speciﬁcity of around 90%5. While different tumour subtypes exhibit typical imaging features on both CT and MRI, there is considerable overlap between certain types e.g., oncocytoma, clear cell and chRCC6. With the exception oftheidentiﬁcation of “macroscopic fat” in lipid-rich angiomyolipomas, there are currently no imaging biomarkers with sufﬁcient accuracy to differentiate benign and malignant tumoursin clinical practice7. A historical reluctance to use biopsy due to the risks, sampling error and a potentially low negative predictive value have meant that rates of benign histology following nephrectomy remain as high as 31%8.\u003Cbr>Therefore, new imaging biomarkers are required that are capable of accurately deﬁning both lesion subtype and metastatic risk. MRI techniques offer valuable insights into tumour biology; particularly diffusion-weighted | (DW) MRI, which can access restricted diffusion via the apparent diffusion coefﬁcient (ADC). DW-MRI has been used to differentiate RCC subgroups9–11 and has proven useful in characterising indeterminate small renal lesions, whether inﬂammatory or malignant, as both may exhibit restricted diffusion12. However, whilst most clinical DW-MRI studies focus on the ADC, it conﬂates multiple physiological parameters, reducing its speciﬁcity for identifying disease mechanisms, with studies showing varying performance in RCC subtype characterisation13–15. The intravoxel incoherent motion (IVIM) model, which separates diffusion in tissue from pseudo-diffusion in bloodvessels, has demonstrated enhanced performance by capturing tumour vascularity16–18, as ﬂow compartments play a signiﬁcant role in microstuctural imaging of the kidney19. While IVIM improves upon the uni-compartmental ADC, it still oversimpliﬁes the complex underlying tissue microstructure.\u003Cbr>The Vascular, Extracell","cbCaicbjFwKonzKj","https://ap.wps.com/l/cbCaicbjFwKonzKj","pdf",2246665,"English","# Dual deep learning approach for noninvasive renal tumour subtyping with VERDICT-MRI\n## Renal cancer background and imaging challenge\n## VERDICT-MRI and study design\n## Model fitting and performance versus ADC/IVIM\n## Reduced b-value protocol and clinical efficiency","[{\"question\":\"What problem does this study address in renal tumour imaging?\",\"answer\":\"Renal tumour subtypes are hard to distinguish using imaging alone, limiting accurate noninvasive characterization and risk assessment.\"},{\"question\":\"How does VERDICT-MRI differ from simpler diffusion models?\",\"answer\":\"VERDICT uses a three-compartment biophysical model that better captures diffusion behaviour in both tumour and healthy tissue compared with ADC-based or simplified approaches.\"},{\"question\":\"What scan-time reduction was achieved and how?\",\"answer\":\"A feature selection method produced a reduced 4 b-value protocol (b = [70, 150, 1000, 2000]), reducing scan time by over 30 minutes to support larger cohort studies.\"}]","Dual deep learning approach for noninvasive renal tumour subtyping with VERDICT-MRI | PDF",1790098458,23,[146,150,154,158],{"page_no":22,"url":147,"width":148,"height":149},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/dbc3606c3fa98e21ec297690ad4adc7f.webp",983,1305,{"page_no":26,"url":151,"width":152,"height":153},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/2e6eb787def17cd294d41d4c2dee8e46.webp",982,1304,{"page_no":30,"url":155,"width":156,"height":157},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/e184efbc72c5e4fb6def10f392b2d5e6.webp",981,1303,{"page_no":33,"url":159,"width":160,"height":161},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/6e852baa0b633c453c2c9b12aedefa43.webp",980,1301]