{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,25]],"date-time":"2026-09-25T14:31:43Z","timestamp":1790346703172,"version":"4.1.0"},"reference-count":18,"publisher":"MDPI AG","issue":"9","license":[{"start":{"date-parts":[[2025,9,22]],"date-time":"2025-09-22T00:00:00Z","timestamp":1758499200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Neurology International"],"abstract":"<jats:p>Background\/Objectives: Multiple sclerosis (MS) is a chronic demyelinating disease where early identification of patients at risk of conversion from clinically isolated syndrome (CIS) to clinically definite MS remains a critical unmet clinical need. Existing machine learning approaches often lack interpretability, limiting clinical trust and adoption. The objective of this research was to develop a novel two-stage machine learning framework with comprehensive explainability to predict CIS-to-MS conversion while addressing demographic bias and interpretability limitations. Methods: A cohort of 177 CIS patients from the National Institute of Neurology and Neurosurgery in Mexico City was analyzed using SeruNet-MS, a two-stage framework that separates demographic baseline risk from clinical risk modification. Stage 1 applied logistic regression to demographic features, while Stage 2 incorporated 25 clinical and symptom features, including MRI lesions, cerebrospinal fluid biomarkers, electrophysiological tests, and symptom characteristics. Patient-level interpretability was achieved through SHAP (SHapley Additive exPlanations) analysis, providing transparent attribution of each factor\u2019s contribution to risk assessment. Results: The two-stage model achieved a ROC-AUC of 0.909, accuracy of 0.806, precision of 0.842, and recall of 0.800, outperforming baseline machine learning methods. Cross-validation confirmed stable performance (0.838 \u00b1 0.095 AUC) with appropriate generalization. SHAP analysis identified periventricular lesions, oligoclonal bands, and symptom complexity as the strongest predictors, with clinical examples illustrating transparent patient-specific risk communication. Conclusions: The two-stage approach effectively mitigates demographic bias by separating non-modifiable factors from actionable clinical findings. SHAP explanations provide clinicians with clear, individualized insights into prediction drivers, enhancing trust and supporting decision making. This framework demonstrates that high predictive performance can be achieved without sacrificing interpretability, representing a significant step forward for explainable AI in MS risk stratification and real-world clinical adoption.<\/jats:p>","DOI":"10.3390\/neurolint17090151","type":"journal-article","created":{"date-parts":[[2025,9,22]],"date-time":"2025-09-22T17:08:17Z","timestamp":1758560897000},"page":"151","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":5,"title":["SeruNet-MS: A Two-Stage Interpretable Framework for Multiple Sclerosis Risk Prediction with SHAP-Based Explainability"],"prefix":"10.3390","volume":"17","author":[{"ORCID":"https:\/\/orcid.org\/0009-0001-2391-9437","authenticated-orcid":false,"given":"Serra","family":"Aksoy","sequence":"first","affiliation":[{"name":"Institute of Computer Science, Ludwig Maximilian University of Munich (LMU), Oettingenstrasse 67, 80538 Munich, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1375-5616","authenticated-orcid":false,"given":"Pinar","family":"Demircioglu","sequence":"additional","affiliation":[{"name":"Department of Mechanical Engineering, Aydin Adnan Menderes University (ADU), Aytepe, Aydin 09010, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9494-5405","authenticated-orcid":false,"given":"Ismail","family":"Bogrekci","sequence":"additional","affiliation":[{"name":"Department of Mechanical Engineering, Aydin Adnan Menderes University (ADU), Aytepe, Aydin 09010, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2025,9,22]]},"reference":[{"key":"ref_1","first-page":"480","article-title":"Optimizing Stroke Classification with Pre-Trained Deep Learning Models","volume":"3","author":"Aksoy","year":"2024","journal-title":"J. Vasc. Dis."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"101593","DOI":"10.1016\/j.nicl.2018.11.003","article-title":"Predicting Conversion from Clinically Isolated Syndrome to Multiple Sclerosis\u2013An Imaging-Based Machine Learning Approach","volume":"21","author":"Zhang","year":"2019","journal-title":"NeuroImage Clin."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"102011","DOI":"10.1016\/j.nicl.2019.102011","article-title":"SVM Recursive Feature Elimination Analyses of Structural Brain MRI Predicts Near-Term Relapses in Patients with Clinically Isolated Syndromes Suggestive of Multiple Sclerosis","volume":"24","author":"Wottschel","year":"2019","journal-title":"NeuroImage Clin."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Ion-M\u0103rgineanu, A., Kocevar, G., Stamile, C., Sima, D.M., Durand-Dubief, F., Van Huffel, S., and Sappey-Marinier, D. (2017). Machine Learning Approach for Classifying Multiple Sclerosis Courses by Combining Clinical Data with Lesion Loads and Magnetic Resonance Metabolic Features. Front. Neurosci., 11.","DOI":"10.3389\/fnins.2017.00398"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"1361","DOI":"10.1007\/s11682-018-9942-9","article-title":"MRI-Based Prediction of Conversion from Clinically Isolated Syndrome to Clinically Definite Multiple Sclerosis Using SVM and Lesion Geometry","volume":"13","author":"Bendfeldt","year":"2019","journal-title":"Brain Imaging Behav."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"250","DOI":"10.1080\/21681163.2017.1356750","article-title":"Deep Learning of Brain Lesion Patterns and User-Defined Clinical and MRI Features for Predicting Conversion to Multiple Sclerosis from Clinically Isolated Syndrome","volume":"7","author":"Yoo","year":"2019","journal-title":"Comput. Methods Biomech. Biomed. Eng. Imaging Vis."},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Daniel, E.C., Tirunagari, S., Batth, K., Windridge, D., and Balla, Y. (2024). Interpretable Machine Learning for Predicting Multiple Sclerosis Conversion from Clinically Isolated Syndrome. medRxiv.","DOI":"10.1101\/2024.07.18.24310578"},{"key":"ref_8","first-page":"2027","article-title":"Machine Learning Algorithms for Early Prediction of Multiple Sclerosis Progression: A Comparative Study","volume":"4","author":"Haouam","year":"2024","journal-title":"Adv. Artif. Intell. Mach. Learn."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Hastie, T., Tibshirani, R., and Friedman, J. (2009). The Elements of Statistical Learning, Springer.","DOI":"10.1007\/978-0-387-84858-7"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"James, G., Witten, D., Hastie, T., and Tibshirani, R. (2021). An Introduction to Statistical Learning: With Applications in R, Springer. Springer Texts in Statistics.","DOI":"10.1007\/978-1-0716-1418-1"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Roy, S., Mincu, D., Proleev, L., Ghate, C., Graves, J.S., Steiner, D.F., Hartsell, F.L., and Heller, K. (2025). Performance of Machine Learning Models for Predicting High-Severity Symptoms in Multiple Sclerosis. Sci. Rep., 15.","DOI":"10.1038\/s41598-024-63888-x"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Rostami, A., Robatjazi, M., Dareyni, A., Ghorbani, A.R., Ganji, O., Siyami, M., and Raoofi, A.R. (2024). Enhancing Classification of Active and Non-Active Lesions in Multiple Sclerosis: Machine Learning Models and Feature Selection Techniques. BMC Med. Imaging, 24.","DOI":"10.1186\/s12880-024-01528-6"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"1336","DOI":"10.1016\/S0140-6736(16)30959-X","article-title":"Diagnosis of Multiple Sclerosis: Progress and Challenges","volume":"389","author":"Brownlee","year":"2017","journal-title":"Lancet"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"162","DOI":"10.1016\/S1474-4422(17)30470-2","article-title":"Diagnosis of Multiple Sclerosis: 2017 Revisions of the McDonald Criteria","volume":"17","author":"Thompson","year":"2018","journal-title":"Lancet Neurol."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"391","DOI":"10.1080\/10408363.2022.2039591","article-title":"Oligoclonal Bands: Clinical Utility and Interpretation Cues","volume":"59","author":"Carta","year":"2022","journal-title":"Crit. Rev. Clin. Lab. Sci."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"1289","DOI":"10.1093\/aje\/kws350","article-title":"Socioeconomic Factors in Childhood and the Risk of Multiple Sclerosis","volume":"177","author":"Nielsen","year":"2013","journal-title":"Am. J. Epidemiol."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"104","DOI":"10.1177\/1352458515579444","article-title":"Level of Education and Multiple Sclerosis Risk after Adjustment for Known Risk Factors: The EnvIMS Study","volume":"22","author":"Riise","year":"2016","journal-title":"Mult. Scler."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"1099","DOI":"10.1016\/S0140-6736(05)71146-6","article-title":"Social Determinants of Health Inequalities","volume":"365","author":"Marmot","year":"2005","journal-title":"Lancet"}],"container-title":["Neurology International"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2035-8377\/17\/9\/151\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,9]],"date-time":"2025-10-09T18:47:12Z","timestamp":1760035632000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2035-8377\/17\/9\/151"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9,22]]},"references-count":18,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2025,9]]}},"alternative-id":["neurolint17090151"],"URL":"https:\/\/doi.org\/10.3390\/neurolint17090151","relation":{},"ISSN":["2035-8377"],"issn-type":[{"value":"2035-8377","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,9,22]]}}}