SeruNet-MS:基于shap可解释性的多发性硬化风险预测的两阶段可解释框架。

IF 3 Q2 CLINICAL NEUROLOGY
Serra Aksoy, Pinar Demircioglu, Ismail Bogrekci
{"title":"SeruNet-MS:基于shap可解释性的多发性硬化风险预测的两阶段可解释框架。","authors":"Serra Aksoy, Pinar Demircioglu, Ismail Bogrekci","doi":"10.3390/neurolint17090151","DOIUrl":null,"url":null,"abstract":"<p><p><b>Background/Objectives:</b> 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. <b>Methods:</b> 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's contribution to risk assessment. <b>Results:</b> 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 ± 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. <b>Conclusions:</b> 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.</p>","PeriodicalId":19130,"journal":{"name":"Neurology International","volume":"17 9","pages":""},"PeriodicalIF":3.0000,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12472785/pdf/","citationCount":"0","resultStr":"{\"title\":\"SeruNet-MS: A Two-Stage Interpretable Framework for Multiple Sclerosis Risk Prediction with SHAP-Based Explainability.\",\"authors\":\"Serra Aksoy, Pinar Demircioglu, Ismail Bogrekci\",\"doi\":\"10.3390/neurolint17090151\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p><b>Background/Objectives:</b> 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. <b>Methods:</b> 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's contribution to risk assessment. <b>Results:</b> 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 ± 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. <b>Conclusions:</b> 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.</p>\",\"PeriodicalId\":19130,\"journal\":{\"name\":\"Neurology International\",\"volume\":\"17 9\",\"pages\":\"\"},\"PeriodicalIF\":3.0000,\"publicationDate\":\"2025-09-22\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12472785/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Neurology International\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.3390/neurolint17090151\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"CLINICAL NEUROLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Neurology International","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.3390/neurolint17090151","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"CLINICAL NEUROLOGY","Score":null,"Total":0}
引用次数: 0

摘要

背景/目的:多发性硬化症(MS)是一种慢性脱髓鞘疾病,早期识别有从临床孤立综合征(CIS)转化为临床明确MS风险的患者仍然是一个关键的未满足的临床需求。现有的机器学习方法往往缺乏可解释性,限制了临床信任和采用。本研究的目的是开发一种具有全面可解释性的新型两阶段机器学习框架,以预测cis到ms的转换,同时解决人口统计学偏差和可解释性限制。方法:使用SeruNet-MS对来自墨西哥城国家神经病学和神经外科研究所的177例CIS患者进行队列分析,SeruNet-MS是一个将人口统计学基线风险与临床风险修改分开的两阶段框架。第1阶段对人口统计学特征进行了逻辑回归,而第2阶段纳入了25个临床和症状特征,包括MRI病变、脑脊液生物标志物、电生理测试和症状特征。通过SHapley加性解释(SHapley Additive explanation)分析实现了患者层面的可解释性,为每个因素对风险评估的贡献提供了透明的归因。结果:两阶段模型的ROC-AUC为0.909,准确率为0.806,精密度为0.842,召回率为0.800,优于基准机器学习方法。交叉验证证实性能稳定(0.838±0.095 AUC),泛化程度适当。SHAP分析确定心室周围病变、寡克隆带和症状复杂性是最强的预测因子,临床实例表明透明的患者特异性风险沟通。结论:两阶段方法通过将不可改变的因素从可操作的临床结果中分离出来,有效地减轻了人口统计学偏差。SHAP解释为临床医生提供了关于预测驱动因素的清晰、个性化的见解,增强了信任并支持了决策制定。该框架表明,在不牺牲可解释性的情况下,可以实现高预测性能,这代表了在MS风险分层和现实世界临床应用中可解释人工智能的重要一步。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
SeruNet-MS: A Two-Stage Interpretable Framework for Multiple Sclerosis Risk Prediction with SHAP-Based Explainability.

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's 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 ± 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.

求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Neurology International
Neurology International CLINICAL NEUROLOGY-
CiteScore
3.70
自引率
3.30%
发文量
69
审稿时长
11 weeks
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信