{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,12]],"date-time":"2026-06-12T19:16:16Z","timestamp":1781291776175,"version":"3.54.1"},"reference-count":37,"publisher":"Public Library of Science (PLoS)","issue":"2","license":[{"start":{"date-parts":[[2026,2,20]],"date-time":"2026-02-20T00:00:00Z","timestamp":1771545600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosdigitalhealth.org"],"crossmark-restriction":false},"short-container-title":["PLOS Digit Health"],"abstract":"<jats:p>This study aimed to develop and evaluate a machine learning pipeline using multiphase contrast-enhanced CT images and clinical data to classify renal tumors as benign, malignant-indolent, or malignant-aggressive, while assessing the contribution of each data source to the classification. In this retrospective study, 448 patients (mean age: 60.7\u2009\u00b1\u200912.6 years, 306 male, 142 female) who underwent nephrectomy and preoperative CECT between June 2008 and July 2018 were included. Tumors were histologically categorized as benign-indolent, malignant-indolent, or malignant-aggressive. Self-supervised feature extraction converted 4-phase CECT images into 512 real-valued features, combined with clinical data and tumor size for classification. Two machine learning classifiers, random forest (RF) and multi-layer perceptron (MLP), were used to predict tumor type. Nested five-fold cross-validation was employed for hyperparameter tuning and model evaluation, and performance was assessed using area under the curve (AUC) analysis. The best-performing models achieved an AUC of 0.90 (95% CI: 0.88\u20130.93) for classifying indolent versus aggressive tumors and 0.76 (95% CI: 0.71\u20130.81) for malignant versus benign tumors. Models incorporating tumor size significantly improved classification accuracy. RF classifiers excelled in distinguishing indolent from aggressive tumors, while MLP classifiers performed better for malignant versus benign classification. The machine learning pipeline demonstrated high accuracy in differentiating aggressive from indolent renal tumors, offering valuable prognostic insights for personalized treatment. Tumor size was a critical factor, complementing CECT images and clinical data. These findings highlight the potential of ML techniques in enhancing renal tumor risk stratification.<\/jats:p>","DOI":"10.1371\/journal.pdig.0001225","type":"journal-article","created":{"date-parts":[[2026,2,20]],"date-time":"2026-02-20T18:34:19Z","timestamp":1771612459000},"page":"e0001225","update-policy":"https:\/\/doi.org\/10.1371\/journal.pdig.corrections_policy","source":"Crossref","is-referenced-by-count":1,"title":["Machine learning based classification of aggressive and malignant renal tumors from multimodal data"],"prefix":"10.1371","volume":"5","author":[{"given":"Mehrnegar","family":"Aminy","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3799-7671","authenticated-orcid":true,"given":"Tejal","family":"Gala","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Agnimitra","family":"Dasgupta","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Serena","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Steven Y.","family":"Cen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"S. J.","family":"Pawan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Inderbir","family":"Gill","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Vinay","family":"Duddalwar","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7668-7713","authenticated-orcid":true,"given":"Assad A.","family":"Oberai","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,2,20]]},"reference":[{"key":"pdig.0001225.ref001","doi-asserted-by":"crossref","DOI":"10.1055\/s-0033-1363840","volume-title":"Surgical management of renal cell carcinoma","author":"LM Krabbe","year":"2014"},{"issue":"36","key":"pdig.0001225.ref002","doi-asserted-by":"crossref","DOI":"10.1200\/JCO.2018.79.1236","article-title":"Imaging Advances in the Management of Kidney Cancer","volume":"36","author":"KM Krajewski","year":"2018","journal-title":"J Clin Oncol"},{"issue":"18","key":"pdig.0001225.ref003","doi-asserted-by":"crossref","first-page":"1331","DOI":"10.1093\/jnci\/djj362","article-title":"Rising incidence of small renal masses: a need to reassess treatment effect","volume":"98","author":"JM Hollingsworth","year":"2006","journal-title":"J Natl Cancer Inst"},{"key":"pdig.0001225.ref004","doi-asserted-by":"crossref","DOI":"10.21037\/atm.2018.12.46","article-title":"Predicting aggressive behavior in small renal tumors prior to treatment","volume":"6","author":"DD Shapiro","year":"2018","journal-title":"Ann Transl Med"},{"issue":"1087","key":"pdig.0001225.ref005","doi-asserted-by":"crossref","first-page":"20180131","DOI":"10.1259\/bjr.20180131","article-title":"CT and MRI of small renal masses","volume":"91","author":"ZJ Wang","year":"2018","journal-title":"Br J Radiol"},{"issue":"4","key":"pdig.0001225.ref006","doi-asserted-by":"crossref","first-page":"973","DOI":"10.1093\/annonc\/mdr362","article-title":"Distribution of metastatic sites in renal cell carcinoma: a population-based analysis","volume":"23","author":"M Bianchi","year":"2012","journal-title":"Ann Oncol"},{"issue":"1","key":"pdig.0001225.ref007","doi-asserted-by":"crossref","first-page":"95","DOI":"10.1016\/j.rcl.2008.01.008","article-title":"Cross-sectional imaging evaluation of renal masses","volume":"46","author":"SR Prasad","year":"2008","journal-title":"Radiol Clin North Am"},{"key":"pdig.0001225.ref008","first-page":"658767","article-title":"Familial renal cancer: molecular genetics and surgical management","volume":"2011","author":"GW Barrisford","year":"2011","journal-title":"Int J Surg Oncol"},{"issue":"3","key":"pdig.0001225.ref009","doi-asserted-by":"crossref","first-page":"620","DOI":"10.21037\/tcr.2017.05.19","article-title":"Renal cell carcinoma: the search for a reliable biomarker","volume":"6","author":"NJ Farber","year":"2017","journal-title":"Transl Cancer Res"},{"issue":"1111","key":"pdig.0001225.ref010","doi-asserted-by":"crossref","first-page":"20200002","DOI":"10.1259\/bjr.20200002","article-title":"Deep learning based classification of solid lipid-poor contrast enhancing renal masses using contrast enhanced CT","volume":"93","author":"A Oberai","year":"2020","journal-title":"Br J Radiol"},{"issue":"1","key":"pdig.0001225.ref011","first-page":"1","article-title":"Surveillance strategies for renal cell carcinoma patients following nephrectomy","volume":"8","author":"AI Chin","year":"2006","journal-title":"Rev Urol"},{"issue":"4","key":"pdig.0001225.ref012","doi-asserted-by":"crossref","first-page":"843","DOI":"10.1016\/S0094-0143(03)00056-9","article-title":"Surveillance after radical or partial nephrectomy for localized renal cell carcinoma and management of recurrent disease","volume":"30","author":"NK Janzen","year":"2003","journal-title":"Urol Clin North Am"},{"issue":"2","key":"pdig.0001225.ref013","doi-asserted-by":"crossref","first-page":"121","DOI":"10.1016\/j.clgc.2012.01.005","article-title":"Prevalence and predictors of benign lesions in renal masses smaller than 7 cm presumed to be renal cell carcinoma","volume":"10","author":"B Akdogan","year":"2012","journal-title":"Clin Genitourin Cancer"},{"key":"pdig.0001225.ref014","article-title":"Association of baseline health and gender with small renal mass pathology","author":"W Bazzi","year":"2014","journal-title":"American Society of Clinical Oncology"},{"issue":"4","key":"pdig.0001225.ref015","doi-asserted-by":"crossref","first-page":"489","DOI":"10.1016\/j.eururo.2018.06.003","article-title":"The Probability of Aggressive Versus Indolent Histology Based on Renal Tumor Size: Implications for Surveillance and Treatment","volume":"74","author":"B Bhindi","year":"2018","journal-title":"Eur Urol"},{"key":"pdig.0001225.ref016","doi-asserted-by":"crossref","first-page":"2217","DOI":"10.1097\/01.ju.0000095475.12515.5e","article-title":"Solid renal tumors: an analysis of pathological features related to tumor size","volume":"170","author":"I Frank","year":"2003","journal-title":"J Urol"},{"issue":"1","key":"pdig.0001225.ref017","doi-asserted-by":"crossref","first-page":"39","DOI":"10.1016\/j.eururo.2011.03.030","article-title":"Active surveillance of small renal masses: progression patterns of early stage kidney cancer","volume":"60","author":"MAS Jewett","year":"2011","journal-title":"Eur Urol"},{"issue":"2","key":"pdig.0001225.ref018","doi-asserted-by":"crossref","first-page":"199","DOI":"10.1097\/JU.0000000000001911","article-title":"Renal Mass and Localized Renal Cancer: Evaluation, Management, and Follow-Up: AUA Guideline: Part I","volume":"206","author":"SC Campbell","year":"2021","journal-title":"J Urol"},{"issue":"6","key":"pdig.0001225.ref019","doi-asserted-by":"crossref","first-page":"668","DOI":"10.1200\/JCO.2016.69.9645","article-title":"Management of Small Renal Masses: American Society of Clinical Oncology Clinical Practice Guideline","volume":"35","author":"A Finelli","year":"2017","journal-title":"J Clin Oncol"},{"issue":"5","key":"pdig.0001225.ref020","doi-asserted-by":"crossref","first-page":"1340","DOI":"10.1016\/j.juro.2015.11.029","article-title":"Diagnostic Accuracy and Risks of Biopsy in the Diagnosis of a Renal Mass Suspicious for Localized Renal Cell Carcinoma: Systematic Review of the Literature","volume":"195","author":"HD Patel","year":"2016","journal-title":"J Urol"},{"issue":"4","key":"pdig.0001225.ref021","doi-asserted-by":"crossref","first-page":"498","DOI":"10.1016\/j.eururo.2018.06.046","article-title":"Prediction of Aggressive Histology: The Ongoing Dilemma of Renal Masses in the \u201cOmics\u201d Era","volume":"74","author":"UA Anele","year":"2018","journal-title":"Eur Urol"},{"key":"pdig.0001225.ref022","doi-asserted-by":"crossref","DOI":"10.1177\/17562872231164803","article-title":"Artificial intelligence and radiomics in evaluation of kidney lesions: a comprehensive literature review","volume":"15","author":"M Ferro","year":"2023","journal-title":"Ther Adv Urol"},{"issue":"2","key":"pdig.0001225.ref023","doi-asserted-by":"crossref","first-page":"292","DOI":"10.1016\/j.tranon.2018.10.012","article-title":"A deep learning-based radiomics model for differentiating benign and malignant renal tumors","volume":"12","author":"L Zhou","year":"2019","journal-title":"Translational Oncology"},{"issue":"4","key":"pdig.0001225.ref024","doi-asserted-by":"crossref","first-page":"553","DOI":"10.1111\/bju.14985","article-title":"Automated differentiation of benign renal oncocytoma and chromophobe renal cell carcinoma on computed tomography using deep learning","volume":"125","author":"A Baghdadi","year":"2020","journal-title":"BJU Int"},{"issue":"10","key":"pdig.0001225.ref025","doi-asserted-by":"crossref","first-page":"1422","DOI":"10.1016\/j.acra.2019.12.015","article-title":"Prediction of Benign and Malignant Solid Renal Masses: Machine Learning-Based CT Texture Analysis","volume":"27","author":"C Erdim","year":"2020","journal-title":"Acad Radiol"},{"issue":"2","key":"pdig.0001225.ref026","doi-asserted-by":"crossref","first-page":"642","DOI":"10.1007\/s00261-022-03735-7","article-title":"Differentiation of benign from malignant solid renal lesions using CT-based radiomics and machine learning: comparison with radiologist interpretation","volume":"48","author":"AL Wentland","year":"2023","journal-title":"Abdom Radiol (NY)"},{"key":"pdig.0001225.ref027","volume-title":"Classification of benign and malignant renal tumors based on CT scans and clinical data using machine learning methods","author":"J Xu","year":"2023"},{"issue":"1","key":"pdig.0001225.ref028","article-title":"Radiologic-Radiomic Machine Learning Models for Differentiation of Benign and Malignant Solid Renal Masses: Comparison With Expert-Level Radiologists","volume":"214","author":"X-Y Sun","year":"2020","journal-title":"AJR Am J Roentgenol"},{"issue":"2","key":"pdig.0001225.ref029","doi-asserted-by":"crossref","first-page":"1254","DOI":"10.1007\/s00330-019-06384-5","article-title":"Radiomics of small renal masses on multiphasic CT: accuracy of machine learning-based classification models for the differentiation of renal cell carcinoma and angiomyolipoma without visible fat","volume":"30","author":"R Yang","year":"2020","journal-title":"Eur Radiol"},{"issue":"6","key":"pdig.0001225.ref030","doi-asserted-by":"crossref","first-page":"4323","DOI":"10.1007\/s00330-022-09384-0","article-title":"Distinguishing common renal cell carcinomas from benign renal tumors based on machine learning: comparing various CT imaging phases, slices, tumor sizes, and ROI segmentation strategies","volume":"33","author":"T Zhou","year":"2023","journal-title":"Eur Radiol"},{"issue":"9","key":"pdig.0001225.ref031","doi-asserted-by":"crossref","first-page":"4278","DOI":"10.1007\/s00261-021-03083-y","article-title":"Evaluation of radiomics and machine learning in identification of aggressive tumor features in renal cell carcinoma (RCC)","volume":"46","author":"S Gurbani","year":"2021","journal-title":"Abdom Radiol (NY)"},{"issue":"1","key":"pdig.0001225.ref032","doi-asserted-by":"crossref","first-page":"300","DOI":"10.1002\/mp.17476","article-title":"Aggressiveness classification of clear cell renal cell carcinoma using registration-independent radiology-pathology correlation learning","volume":"52","author":"I Bhattacharya","year":"2025","journal-title":"Med Phys"},{"key":"pdig.0001225.ref033","article-title":"A simple framework for contrastive learning of visual representations","author":"T Chen","year":"2020"},{"issue":"2","key":"pdig.0001225.ref034","doi-asserted-by":"crossref","first-page":"1011","DOI":"10.1007\/s00330-020-07158-0","article-title":"Shape and texture-based radiomics signature on CT effectively discriminates benign from malignant renal masses","volume":"31","author":"FY Yap","year":"2021","journal-title":"Eur Radiol"},{"key":"pdig.0001225.ref035","doi-asserted-by":"crossref","first-page":"121","DOI":"10.1016\/j.urology.2017.12.018","article-title":"Quantitative Contour Analysis as an Image-based Discriminator Between Benign and Malignant Renal Tumors","volume":"114","author":"FY Yap","year":"2018","journal-title":"Urology"},{"issue":"1","key":"pdig.0001225.ref036","doi-asserted-by":"crossref","first-page":"32","DOI":"10.1002\/1097-0142(1950)3:1<32::AID-CNCR2820030106>3.0.CO;2-3","article-title":"Index for rating diagnostic tests","volume":"3","author":"WJ Youden","year":"1950","journal-title":"Cancer"},{"issue":"1","key":"pdig.0001225.ref037","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1175\/1520-0493(1950)078<0001:VOFEIT>2.0.CO;2","article-title":"Verification of Forecasts Expressed in Terms of Probability","volume":"78","author":"GW Brier","year":"1950","journal-title":"Mon Wea Rev"}],"container-title":["PLOS Digital Health"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0001225","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,20]],"date-time":"2026-02-20T18:34:25Z","timestamp":1771612465000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pdig.0001225"}},"subtitle":[],"editor":[{"given":"Satyananda","family":"Kashyap","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,2,20]]},"references-count":37,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2026,2,20]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pdig.0001225","relation":{},"ISSN":["2767-3170"],"issn-type":[{"value":"2767-3170","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,20]]}}}