{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T03:28:18Z","timestamp":1791430098749,"version":"4.3.3"},"reference-count":47,"publisher":"MDPI AG","issue":"9","license":[{"start":{"date-parts":[[2026,4,28]],"date-time":"2026-04-28T00:00:00Z","timestamp":1777334400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Ministry of Economic Development of the Russian Federation","award":["139-15-2025-004"],"award-info":[{"award-number":["139-15-2025-004"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Cancers"],"abstract":"<jats:p>Background: Prostate cancer (PCa) is one of the most prevalent malignancies in men, and active surveillance (AS) is the recommended management strategy for low- and favourable intermediate-risk disease. Predicting which patients will progress during AS remains a clinical challenge. MRI-derived radiomics has shown promise for risk stratification, but conventional machine learning approaches treat radiomic features as independent variables and may not capture the complex inter-feature dependencies within imaging data. This study evaluates the application of Synolitic Graph Neural Networks (SGNNs) to baseline MRI-derived radiomic features for predicting prostate cancer progression on active surveillance. Methods: We studied 343 AS patients (73 progressors, 270 non-progressors) from a single-centre cohort prospectively enrolled between 2013 and 2019 and retrospectively analysed. Seventy-two radiomic features were extracted from baseline 3T MRI (T2-weighted imaging and apparent diffusion coefficient maps), together with three clinical variables (prostate volume, PSA, PSA density). The SGNN pipeline transformed each patient\u2019s feature profile into a weighted graph encoding pairwise feature relationships via logistic regression classifiers trained within each cross-validation fold. GCN and GATv2 architectures were evaluated with multiple sparsification strategies and compared against Gradient Boosting, SVM, Random Forest, and logistic regression using 5-fold stratified cross-validation. Results: Among conventional methods, Gradient Boosting achieved the highest ROC-AUC (0.634 \u00b1 0.080). The SGNN pipeline with GATv2, confidence-based sparsification (p = 0.8), and extended node features incorporating graph centrality measures achieved the best performance (ROC-AUC = 0.699 \u00b1 0.044), an absolute improvement of 0.065 over the best conventional method. The addition of topological node features consistently improved performance by 3\u20135% across configurations. GATv2 outperformed GCN in matched settings. Conclusions: As a proof of concept, the SGNN framework achieved the highest mean ROC-AUC among the evaluated single-timepoint approaches, though results require confirmation in independent external cohorts. By encoding inter-feature relationships as patient-specific graphs, SGNN offers a complementary modelling paradigm for radiomic data in clinical oncology. Future work should incorporate longitudinal data and graph explainability methods.<\/jats:p>","DOI":"10.3390\/cancers18091389","type":"journal-article","created":{"date-parts":[[2026,4,28]],"date-time":"2026-04-28T07:49:36Z","timestamp":1777362576000},"page":"1389","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Synolitic Graph Neural Networks for MRI-Derived Radiomic-Based Prediction of Prostate Cancer Progression on Active Surveillance"],"prefix":"10.3390","volume":"18","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-1169-5149","authenticated-orcid":false,"given":"Mikhail I.","family":"Krivonosov","sequence":"first","affiliation":[{"name":"Research Center in Artificial Intelligence, Institute of Information Technologies, Mathematics and Mechanics, Lobachevsky State University, Nizhny Novgorod 603022, Russia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6398-6549","authenticated-orcid":false,"given":"Arseniy","family":"Trukhanov","sequence":"additional","affiliation":[{"name":"Mriya Life Institute, National Academy of Active Longevity, Moscow 124489, Russia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Nikita","family":"Sushentsev","sequence":"additional","affiliation":[{"name":"Department of Radiology, University of Cambridge School of Clinical Medicine, Cambridge CB2 0QQ, UK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tristan","family":"Barrett","sequence":"additional","affiliation":[{"name":"Department of Radiology, University of Cambridge School of Clinical Medicine, Cambridge CB2 0QQ, UK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7540-1130","authenticated-orcid":false,"given":"Alexey","family":"Zaikin","sequence":"additional","affiliation":[{"name":"Research Center in Artificial Intelligence, Institute of Information Technologies, Mathematics and Mechanics, Lobachevsky State University, Nizhny Novgorod 603022, Russia"},{"name":"Institute for Cognitive Neuroscience, University Higher School of Economics, Moscow 101000, Russia"},{"name":"Department of Mathematics and Women\u2019s Cancer, University College London, London WC1E 6BT, UK"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,4,28]]},"reference":[{"key":"ref_1","first-page":"229","article-title":"Global Cancer Statistics 2022: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries","volume":"74","author":"Bray","year":"2024","journal-title":"CA Cancer J. Clin."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"1795","DOI":"10.1038\/bjc.2011.430","article-title":"Cancer incidence in the United Kingdom: Projections to the year 2030","volume":"105","author":"Mistry","year":"2011","journal-title":"Br. J. Cancer"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"320","DOI":"10.1016\/j.eururo.2019.10.030","article-title":"Ten-year mortality, disease progression, and treatment-related side effects in men with localised prostate cancer from the ProtecT randomised controlled trial","volume":"77","author":"Neal","year":"2020","journal-title":"Eur. Urol."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"132","DOI":"10.1056\/NEJMoa1615869","article-title":"Follow-up of prostatectomy versus observation for early prostate cancer","volume":"377","author":"Wilt","year":"2017","journal-title":"N. Engl. J. Med."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"126","DOI":"10.1200\/JCO.2009.24.2180","article-title":"Clinical results of long-term follow-up of a large, active surveillance cohort with localized prostate cancer","volume":"28","author":"Klotz","year":"2010","journal-title":"J. Clin. Oncol."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"203","DOI":"10.1056\/NEJMoa1113162","article-title":"Radical prostatectomy versus observation for localized prostate cancer","volume":"367","author":"Wilt","year":"2012","journal-title":"N. Engl. J. Med."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"2809","DOI":"10.21037\/tau-20-1370","article-title":"Active Surveillance for Prostate Cancer","volume":"10","author":"Shill","year":"2021","journal-title":"Transl. Androl. Urol."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"2084","DOI":"10.1001\/jama.2024.6695","article-title":"Long-Term Outcomes in Patients Using Protocol-Directed Active Surveillance for Prostate Cancer","volume":"331","author":"Newcomb","year":"2024","journal-title":"JAMA"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"597","DOI":"10.1016\/j.eururo.2012.11.005","article-title":"Active Surveillance for Low-Risk Prostate Cancer Worldwide: The PRIAS Study","volume":"63","author":"Bul","year":"2013","journal-title":"Eur. Urol."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"954","DOI":"10.1016\/j.eururo.2016.06.007","article-title":"A Decade of Active Surveillance in the PRIAS Study","volume":"70","author":"Bokhorst","year":"2016","journal-title":"Eur. Urol."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"131","DOI":"10.2214\/AJR.16.16355","article-title":"The Emerging Role of MRI in Prostate Cancer Active Surveillance and Ongoing Challenges","volume":"208","author":"Barrett","year":"2017","journal-title":"Am. J. Roentgenol."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"523","DOI":"10.1016\/j.eururo.2018.10.025","article-title":"Reasons for Discontinuing Active Surveillance: Assessment of 21 Centres in 12 Countries in the Movember GAP3 Consortium","volume":"75","author":"Ji","year":"2019","journal-title":"Eur. Urol."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"758","DOI":"10.1111\/bju.14800","article-title":"Using prognosis to guide inclusion criteria, define standardised endpoints and stratify follow-up in active surveillance for prostate cancer","volume":"124","author":"Gnanapragasam","year":"2019","journal-title":"BJU Int."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"78","DOI":"10.1016\/j.eururo.2019.06.023","article-title":"Prostate magnetic resonance imaging, with or without MRI-targeted biopsy, and systematic biopsy for detecting prostate cancer: A Cochrane systematic review and meta-analysis","volume":"77","author":"Drost","year":"2020","journal-title":"Eur. Urol."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"443","DOI":"10.1016\/j.eururo.2020.03.035","article-title":"Five-year outcomes of magnetic resonance imaging-based active surveillance for prostate cancer","volume":"78","author":"Stavrinides","year":"2020","journal-title":"Eur. Urol."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"2696","DOI":"10.1007\/s00330-020-07336-0","article-title":"MRI-derived PRECISE scores for predicting pathologically-confirmed radiological progression in prostate cancer patients on active surveillance","volume":"31","author":"Caglic","year":"2021","journal-title":"Eur. Radiol."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"59","DOI":"10.1111\/bju.14166","article-title":"Progression and treatment rates using an active surveillance protocol incorporating image-guided baseline biopsies and multiparametric MRI monitoring","volume":"122","author":"Thurtle","year":"2018","journal-title":"BJU Int."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"809","DOI":"10.1016\/j.eururo.2013.03.025","article-title":"Value of targeted prostate biopsy using MR-US fusion in men with prior negative biopsy and elevated PSA","volume":"65","author":"Sonn","year":"2014","journal-title":"Eur. Urol."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"648","DOI":"10.1016\/j.eururo.2016.06.011","article-title":"Reporting Magnetic Resonance Imaging in Men on Active Surveillance for Prostate Cancer: The PRECISE Recommendations","volume":"71","author":"Moore","year":"2017","journal-title":"Eur. Urol."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"549","DOI":"10.1016\/j.eururo.2021.05.001","article-title":"Reliability of Serial Prostate MRI to Detect Prostate Cancer Progression During Active Surveillance","volume":"80","author":"Rajwa","year":"2021","journal-title":"Eur. Urol."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"235","DOI":"10.1038\/s41391-018-0111-4","article-title":"The role of prostate imaging reporting and data system score in Gleason 3+3 active surveillance candidates enrollment: A diagnostic meta-analysis","volume":"22","author":"Zhai","year":"2019","journal-title":"Prostate Cancer Prostatic Dis."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"109095","DOI":"10.1016\/j.ejrad.2020.109095","article-title":"Prostate MRI radiomics: A systematic review and radiomic quality score assessment","volume":"129","author":"Stanzione","year":"2020","journal-title":"Eur. J. Radiol."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"563","DOI":"10.1148\/radiol.2015151169","article-title":"Radiomics: Images Are More than Pictures, They Are Data","volume":"278","author":"Gillies","year":"2016","journal-title":"Radiology"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"1313","DOI":"10.1007\/s00330-019-06488-y","article-title":"Multiparametric MRI and auto-fixed volume of interest-based radiomics signature for clinically significant peripheral zone prostate cancer","volume":"30","author":"Bleker","year":"2020","journal-title":"Eur. Radiol."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"4223","DOI":"10.1007\/s00261-020-02678-1","article-title":"A radiomics machine learning-based redefining score robustly identifies clinically significant prostate cancer in equivocal PI-RADS score 3 lesions","volume":"45","author":"Hou","year":"2020","journal-title":"Abdom. Radiol."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"16","DOI":"10.1016\/j.ejrad.2019.03.010","article-title":"Multi-parametric MRI-based radiomics signature for discriminating between clinically significant and insignificant prostate cancer","volume":"115","author":"Min","year":"2019","journal-title":"Eur. J. Radiol."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"1239","DOI":"10.1002\/jmri.27138","article-title":"Radiomics based on MRI as a biomarker to guide therapy by predicting upgrading of prostate cancer from biopsy to radical prostatectomy","volume":"52","author":"Zhang","year":"2020","journal-title":"J. Magn. Reson. Imaging"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"12917","DOI":"10.1038\/s41598-021-92341-6","article-title":"MRI-derived radiomics model for baseline prediction of prostate cancer progression on active surveillance","volume":"11","author":"Sushentsev","year":"2021","journal-title":"Sci. Rep."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"680","DOI":"10.1007\/s00330-021-08151-x","article-title":"Comparative Performance of MRI-Derived PRECISE Scores and Delta-Radiomics Models for the Prediction of Prostate Cancer Progression in Patients on Active Surveillance","volume":"32","author":"Sushentsev","year":"2022","journal-title":"Eur. Radiol."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"3792","DOI":"10.1007\/s00330-023-09438-x","article-title":"Time Series Radiomics for the Prediction of Prostate Cancer Progression in Patients on Active Surveillance","volume":"33","author":"Sushentsev","year":"2023","journal-title":"Eur. Radiol."},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Vershinina, O., Sushentsev, N., Zaikin, A., Blyuss, O., Barrett, T., and Ivanchenko, M. (2025). Improving the Potential for Predicting Prostate Cancer Progression in Patients on Active Surveillance Using Explainable Artificial Intelligence. Cancers, 17.","DOI":"10.3390\/cancers17223598"},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"4","DOI":"10.1109\/TNNLS.2020.2978386","article-title":"A Comprehensive Survey on Graph Neural Networks","volume":"32","author":"Wu","year":"2021","journal-title":"IEEE Trans. Neural Netw. Learn. Syst."},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Ahmedt-Aristizabal, D., Armin, M.A., Denman, S., Fookes, C., and Petersson, L. (2021). Graph-Based Deep Learning for Medical Diagnosis and Analysis: Past, Present and Future. Sensors, 21.","DOI":"10.3390\/s21144758"},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Krivonosov, M., Nazarenko, T., Ushakov, V., and Zaikin, A. (2025). Analysis of Multidimensional Clinical and Physiological Data with Synolitical Graph Neural Networks. Technologies, 13.","DOI":"10.3390\/technologies13010013"},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Ney, A., Nen\u00e9, N.R., Sedlak, E., Acedo, P., Blyuss, O., Whitwell, H.J., Costello, E., Gentry-Maharaj, A., Williams, N.R., and Menon, U. (2024). Identification of a serum proteomic biomarker panel using diagnosis specific ensemble learning and symptoms for early pancreatic cancer detection. PLoS Comput. Biol., 20.","DOI":"10.1371\/journal.pcbi.1012408"},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"10","DOI":"10.1038\/s43856-023-00237-5","article-title":"Serum biomarker-based early detection of pancreatic ductal adenocarcinomas with ensemble learning","volume":"3","author":"Ney","year":"2023","journal-title":"Commun. Med."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"20190931","DOI":"10.1259\/bjr.20190929","article-title":"The effect of capped biparametric magnetic resonance imaging slots on weekly prostate cancer imaging workload","volume":"93","author":"Sushentsev","year":"2020","journal-title":"Br. J. Radiol."},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"1116","DOI":"10.1016\/j.neuroimage.2006.01.015","article-title":"User-guided 3D active contour segmentation of anatomical structures: Significantly improved efficiency and reliability","volume":"31","author":"Yushkevich","year":"2006","journal-title":"NeuroImage"},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"e104","DOI":"10.1158\/0008-5472.CAN-17-0339","article-title":"Computational Radiomics System to Decode the Radiographic Phenotype","volume":"77","author":"Fedorov","year":"2017","journal-title":"Cancer Res."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"328","DOI":"10.1148\/radiol.2020191145","article-title":"The Image Biomarker Standardization Initiative: Standardized Quantitative Radiomics for High-Throughput Image-Based Phenotyping","volume":"295","author":"Zwanenburg","year":"2020","journal-title":"Radiology"},{"key":"ref_41","unstructured":"Kipf, T.N., and Welling, M. (2017, January 24\u201326). Semi-Supervised Classification with Graph Convolutional Networks. Proceedings of the International Conference on Learning Representations (ICLR), Toulon, France."},{"key":"ref_42","unstructured":"Brody, S., Alon, U., and Yahav, E. (2022, January 25\u201329). How Attentive are Graph Attention Networks?. Proceedings of the International Conference on Learning Representations (ICLR), Virtual."},{"key":"ref_43","doi-asserted-by":"crossref","unstructured":"Akiba, T., Sano, S., Yanase, T., Ohta, T., and Koyama, M. (2019, January 4\u20138). Optuna: A Next-generation Hyperparameter Optimization Framework. Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, Anchorage, AK, USA.","DOI":"10.1145\/3292500.3330701"},{"key":"ref_44","unstructured":"Wallach, H., Larochelle, H., Beygelzimer, A., d\u2019Alch\u00e9-Buc, F., Fox, E., and Garnett, R. (2019). GNNExplainer: Generating Explanations for Graph Neural Networks. Advances in Neural Information Processing Systems, Curran Associates, Inc."},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"3","DOI":"10.21873\/invivo.13569","article-title":"Oncological Outcomes in Men with Favorable Intermediate Risk Prostate Cancer Enrolled in Active Surveillance","volume":"38","author":"Pepe","year":"2024","journal-title":"In Vivo"},{"key":"ref_46","doi-asserted-by":"crossref","unstructured":"Rosa, E., Vaccaro, M., Placidi, E., D\u2019Andrea, M.L., Liporace, F., Natali, G.L., Secinaro, A., and Napolitano, A. (2025). Quantum Neural Networks in Magnetic Resonance Imaging: Advancing Diagnostic Precision Through Emerging Computational Paradigms. Computers, 14.","DOI":"10.3390\/computers14120529"},{"key":"ref_47","doi-asserted-by":"crossref","first-page":"10248","DOI":"10.1038\/s41598-020-66110-w","article-title":"Performance comparison of modified ComBat for harmonization of radiomic features for multicenter studies","volume":"10","author":"Masson","year":"2020","journal-title":"Sci. Rep."}],"container-title":["Cancers"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2072-6694\/18\/9\/1389\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,28]],"date-time":"2026-04-28T08:06:53Z","timestamp":1777363613000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2072-6694\/18\/9\/1389"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,28]]},"references-count":47,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2026,5]]}},"alternative-id":["cancers18091389"],"URL":"https:\/\/doi.org\/10.3390\/cancers18091389","relation":{},"ISSN":["2072-6694"],"issn-type":[{"value":"2072-6694","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,28]]}}}