Wenqi Hu, Qiaohui Du, Lisi Wei, Dawei Wang, Guang Zhang
{"title":"基于海马MRI分割、特征融合和图建模的新型多任务学习用于阿尔茨海默氏症分期分类。","authors":"Wenqi Hu, Qiaohui Du, Lisi Wei, Dawei Wang, Guang Zhang","doi":"10.1186/s40001-025-03204-z","DOIUrl":null,"url":null,"abstract":"<p><strong>Objective: </strong>To develop and validate a comprehensive and interpretable framework for multi-class classification of Alzheimer's disease (AD) progression stages based on hippocampal MRI, integrating radiomic, deep, and clinical features.</p><p><strong>Materials and methods: </strong>This retrospective multi-center study included 2956 patients across four AD stages (Non-Demented, Very Mild Demented, Mild Demented, Moderate Demented). T1-weighted MRI scans were processed through a standardized pipeline involving hippocampal segmentation using four models (U-Net, nnU-Net, Swin-UNet, MedT). Radiomic features (n = 215) were extracted using the SERA platform, and deep features (n = 256) were learned using an LSTM network with attention applied to hippocampal slices. Fused features were harmonized with ComBat and filtered by ICC (≥ 0.75), followed by LASSO-based feature selection. Classification was performed using five machine learning models, including Logistic Regression (LR), Support Vector Machine (SVM), Random Forest (RF), Multilayer Perceptron (MLP), and eXtreme Gradient Boosting (XGBoost). Model interpretability was addressed using SHAP, and a nomogram and decision curve analysis (DCA) were developed. Additionally, an end-to-end 3D CNN-LSTM model and two transformer-based benchmarks (Vision Transformer, Swin Transformer) were trained for comparative evaluation.</p><p><strong>Results: </strong>MedT achieved the best hippocampal segmentation (Dice = 92.03% external). Fused features yielded the highest classification performance with XGBoost (external accuracy = 92.8%, AUC = 94.2%). SHAP identified MMSE, hippocampal volume, and APOE ε4 as top contributors. The nomogram accurately predicted early-stage AD with clinical utility confirmed by DCA. The end-to-end model performed acceptably (AUC = 84.0%) but lagged behind the fused pipeline. Statistical tests confirmed significant performance advantages for feature fusion and MedT-based segmentation.</p><p><strong>Conclusions: </strong>This study demonstrates that integrating radiomics, deep learning, and clinical data from hippocampal MRI enables accurate and interpretable classification of AD stages. The proposed framework is robust, generalizable, and clinically actionable, representing a scalable solution for AD diagnostics.</p>","PeriodicalId":11949,"journal":{"name":"European Journal of Medical Research","volume":"30 1","pages":"910"},"PeriodicalIF":3.4000,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12482563/pdf/","citationCount":"0","resultStr":"{\"title\":\"Novel multi-task learning for Alzheimer's stage classification using hippocampal MRI segmentation, feature fusion, and nomogram modeling.\",\"authors\":\"Wenqi Hu, Qiaohui Du, Lisi Wei, Dawei Wang, Guang Zhang\",\"doi\":\"10.1186/s40001-025-03204-z\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><strong>Objective: </strong>To develop and validate a comprehensive and interpretable framework for multi-class classification of Alzheimer's disease (AD) progression stages based on hippocampal MRI, integrating radiomic, deep, and clinical features.</p><p><strong>Materials and methods: </strong>This retrospective multi-center study included 2956 patients across four AD stages (Non-Demented, Very Mild Demented, Mild Demented, Moderate Demented). T1-weighted MRI scans were processed through a standardized pipeline involving hippocampal segmentation using four models (U-Net, nnU-Net, Swin-UNet, MedT). Radiomic features (n = 215) were extracted using the SERA platform, and deep features (n = 256) were learned using an LSTM network with attention applied to hippocampal slices. Fused features were harmonized with ComBat and filtered by ICC (≥ 0.75), followed by LASSO-based feature selection. Classification was performed using five machine learning models, including Logistic Regression (LR), Support Vector Machine (SVM), Random Forest (RF), Multilayer Perceptron (MLP), and eXtreme Gradient Boosting (XGBoost). Model interpretability was addressed using SHAP, and a nomogram and decision curve analysis (DCA) were developed. Additionally, an end-to-end 3D CNN-LSTM model and two transformer-based benchmarks (Vision Transformer, Swin Transformer) were trained for comparative evaluation.</p><p><strong>Results: </strong>MedT achieved the best hippocampal segmentation (Dice = 92.03% external). Fused features yielded the highest classification performance with XGBoost (external accuracy = 92.8%, AUC = 94.2%). SHAP identified MMSE, hippocampal volume, and APOE ε4 as top contributors. The nomogram accurately predicted early-stage AD with clinical utility confirmed by DCA. The end-to-end model performed acceptably (AUC = 84.0%) but lagged behind the fused pipeline. Statistical tests confirmed significant performance advantages for feature fusion and MedT-based segmentation.</p><p><strong>Conclusions: </strong>This study demonstrates that integrating radiomics, deep learning, and clinical data from hippocampal MRI enables accurate and interpretable classification of AD stages. The proposed framework is robust, generalizable, and clinically actionable, representing a scalable solution for AD diagnostics.</p>\",\"PeriodicalId\":11949,\"journal\":{\"name\":\"European Journal of Medical Research\",\"volume\":\"30 1\",\"pages\":\"910\"},\"PeriodicalIF\":3.4000,\"publicationDate\":\"2025-09-29\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12482563/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"European Journal of Medical Research\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1186/s40001-025-03204-z\",\"RegionNum\":3,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"MEDICINE, RESEARCH & EXPERIMENTAL\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"European Journal of Medical Research","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1186/s40001-025-03204-z","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"MEDICINE, RESEARCH & EXPERIMENTAL","Score":null,"Total":0}
Novel multi-task learning for Alzheimer's stage classification using hippocampal MRI segmentation, feature fusion, and nomogram modeling.
Objective: To develop and validate a comprehensive and interpretable framework for multi-class classification of Alzheimer's disease (AD) progression stages based on hippocampal MRI, integrating radiomic, deep, and clinical features.
Materials and methods: This retrospective multi-center study included 2956 patients across four AD stages (Non-Demented, Very Mild Demented, Mild Demented, Moderate Demented). T1-weighted MRI scans were processed through a standardized pipeline involving hippocampal segmentation using four models (U-Net, nnU-Net, Swin-UNet, MedT). Radiomic features (n = 215) were extracted using the SERA platform, and deep features (n = 256) were learned using an LSTM network with attention applied to hippocampal slices. Fused features were harmonized with ComBat and filtered by ICC (≥ 0.75), followed by LASSO-based feature selection. Classification was performed using five machine learning models, including Logistic Regression (LR), Support Vector Machine (SVM), Random Forest (RF), Multilayer Perceptron (MLP), and eXtreme Gradient Boosting (XGBoost). Model interpretability was addressed using SHAP, and a nomogram and decision curve analysis (DCA) were developed. Additionally, an end-to-end 3D CNN-LSTM model and two transformer-based benchmarks (Vision Transformer, Swin Transformer) were trained for comparative evaluation.
Results: MedT achieved the best hippocampal segmentation (Dice = 92.03% external). Fused features yielded the highest classification performance with XGBoost (external accuracy = 92.8%, AUC = 94.2%). SHAP identified MMSE, hippocampal volume, and APOE ε4 as top contributors. The nomogram accurately predicted early-stage AD with clinical utility confirmed by DCA. The end-to-end model performed acceptably (AUC = 84.0%) but lagged behind the fused pipeline. Statistical tests confirmed significant performance advantages for feature fusion and MedT-based segmentation.
Conclusions: This study demonstrates that integrating radiomics, deep learning, and clinical data from hippocampal MRI enables accurate and interpretable classification of AD stages. The proposed framework is robust, generalizable, and clinically actionable, representing a scalable solution for AD diagnostics.
期刊介绍:
European Journal of Medical Research publishes translational and clinical research of international interest across all medical disciplines, enabling clinicians and other researchers to learn about developments and innovations within these disciplines and across the boundaries between disciplines. The journal publishes high quality research and reviews and aims to ensure that the results of all well-conducted research are published, regardless of their outcome.