What question did this study set out to answer?

This research aims to determine how well machine learning models using multimodal MRI can differentiate between patients with and without depression in cerebral small vessel disease.

April 3, 2026BMC Medical Imaging1 citationsOpen Access

Multimodal MRI and machine learning for identifying depression in cerebral small vessel disease: a multicenter study

GLGuihan LinWenzhou Medical University WCWeiyue ChenWenzhou Medical University YCYongjun ChenWenzhou Medical University

Key Points

This research aims to determine how well machine learning models using multimodal MRI can differentiate between patients with and without depression in cerebral small vessel disease.
Conducted a retrospective study with 198 participants across three centers.
Utilized structural, functional, and diffusion tensor imaging to extract relevant features.
Constructed and evaluated machine learning models using nine classifiers.
Analyzed model efficacy using receiver operating characteristic (ROC) curves and SHapley additive explanations (SHAP).
Identified twelve key features from multimodal MRI that contribute to model accuracy.
The eXtreme Gradient Boosting (XGBoost) classifier demonstrated the highest diagnostic performance.
Achieved areas under the ROC curves of 0.958, 0.893, and 0.917 for training and validation cohorts.
Provided accuracies of 0.929, 0.871, and 0.853 across the respective cohorts.
Key features contributing to model accuracy included altered brain activity and microstructural changes.

Abstract

This study sought to evaluate the effectiveness of a machine learning (ML) model utilizing multimodal MRI in distinguishing cerebral small vessel disease (CSVD) patients with depression (CSVD + D) from those without depression (CSVD − D). This retrospective study involved 198 participants from three centers, who were divided into training (n = 113; CSVD + D = 56, CSVD − D = 57), external validation 1 (n = 85; CSVD + D = 31, CSVD − D = 54), and 2 (n = 102; CSVD + D = 39, CSVD − D = 63) cohorts. Structural, functional, and diffusion tensor imaging was used to extract features, which were utilized to construct ML models based on nine ML classifiers. The efficacy of the models was evaluated through the receiver operating characteristic (ROC) analysis. SHapley additive explanations (SHAP) analysis provided deep insights into the model’s interpretability. Twelve features from multimodal MRI were finally identified. The eXtreme Gradient Boosting (XGBoost) classifier performed the best. The XGBoost-based multimodal model integrating features from all three modalities achieved high diagnostic performance, with areas under the ROC curves of 0.958, 0.893, and 0.917 in the training, external validation 1 and 2 cohorts, respectively, and accuracies of 0.929, 0.871, and 0.853. The SHAP results revealed that the key contributing features included elevated amplitude of low-frequency fluctuations in the superior frontal gyrus and reduced fractional anisotropy in the default mode network. Integrating multimodal MRI and ML may improve the classification performance for identifying CSVD + D, suggesting the potential value of multimodal imaging markers in the assessment of CSVD-related depressive symptoms. Not applicable. Cerebral small vessel disease with depression (CSVD + D) correlates with greater cognitive and emotional impairment, underscoring the need for early identification. Multimodal MRI features provide complementary information that enhances the identification of CSVD + D patients. A multimodal MRI model based on the eXtreme Gradient Boosting classifier improves diagnostic accuracy for CSVD + D compared to other machine learning models.

问 AI

Bookmark

View Full Paper