Classification of Major Depressive Disorder Using Vertex-Wise Brain Sulcal Depth, Curvature, and Thickness with a Deep and a Shallow Learning Model
Abstract
Major depressive disorder (MDD) is a complex psychiatric disorder that affects the lives of hundreds of millions of individuals around the globe. Even today, researchers debate if morphological alterations in the brain are linked to MDD, likely due to the heterogeneity of this disorder. The application of deep learning tools to neuroimaging data, capable of capturing complex non-linear patterns, has the potential to provide diagnostic and predictive biomarkers for MDD. However, previous attempts to demarcate MDD patients and healthy controls (HC) based on segmented cortical features via linear machine learning approaches have reported low accuracies. Here, we used globally representative data from the ENIGMA-MDD working group containing 7,012 participants from 30 sites (N=2,772 MDD and N=4,240 HC), which allows a comprehensive analysis with generalizable results. Based on the hypothesis that integration of vertex-wise cortical features can improve classification performance, we evaluated the classification of a DenseNet and a Support Vector Machine (SVM), with the expectation that the former would outperform the latter. We found that both classifiers exhibited close to chance performance (balanced accuracy DenseNet: 51%; SVM: 53%), when estimated on unseen sites. Slightly higher classification performance (balanced accuracy DenseNet: 58%; SVM: 55%) was found when the cross-validation folds contained subjects from all sites, indicating site effect. In conclusion, the integration of vertex-wise morphometric features and the use of the non-linear classifier did not lead to the differentiability between MDD and HC. Our results support the notion that MDD classification on this combination of such features and classifiers is unfeasible. Perhaps more sophisticated integration of multimodal information may lead to a higher performance in this diagnostic task.
Keywords
Cite
@article{arxiv.2311.11046,
title = {Classification of Major Depressive Disorder Using Vertex-Wise Brain Sulcal Depth, Curvature, and Thickness with a Deep and a Shallow Learning Model},
author = {Roberto Goya-Maldonado and Tracy Erwin-Grabner and Ling-Li Zeng and Christopher R. K. Ching and Andre Aleman and Alyssa R. Amod and Zeynep Basgoze and Francesco Benedetti and Bianca Besteher and Katharina Brosch and Robin Bülow and Romain Colle and Colm G. Connolly and Emmanuelle Corruble and Baptiste Couvy-Duchesne and Kathryn Cullen and Udo Dannlowski and Christopher G. Davey and Annemiek Dols and Jan Ernsting and Jennifer W. Evans and Lukas Fisch and Paola Fuentes-Claramonte and Ali Saffet Gonul and Ian H. Gotlib and Hans J. Grabe and Nynke A. Groenewold and Dominik Grotegerd and Tim Hahn and J. Paul Hamilton and Laura K. M. Han and Ben J. Harrison and Tiffany C. Ho and Neda Jahanshad and Alec J. Jamieson and Andriana Karuk and Tilo Kircher and Bonnie Klimes-Dougan and Sheri-Michelle Koopowitz and Thomas Lancaster and Ramona Leenings and Meng Li and David E. J. Linden and Frank P. MacMaster and David M. A. Mehler and Susanne Meinert and Elisa Melloni and Bryon A. Mueller and Benson Mwangi and Igor Nenadić and Amar Ojha and Yasumasa Okamoto and Mardien L. Oudega and Brenda W. J. H. Penninx and Sara Poletti and Edith Pomarol-Clotet and Maria J. Portella and Elena Pozzi and Joaquim Radua and Elena Rodríguez-Cano and Matthew D. Sacchet and Raymond Salvador and Anouk Schrantee and Kang Sim and Jair C. Soares and Aleix Solanes and Dan J. Stein and Frederike Stein and Aleks Stolicyn and Sophia I. Thomopoulos and Yara J. Toenders and Aslihan Uyar-Demir and Eduard Vieta and Yolanda Vives-Gilabert and Henry Völzke and Martin Walter and Heather C. Whalley and Sarah Whittle and Nils Winter and Katharina Wittfeld and Margaret J. Wright and Mon-Ju Wu and Tony T. Yang and Carlos Zarate and Dick J. Veltman and Lianne Schmaal and Paul M. Thompson},
journal= {arXiv preprint arXiv:2311.11046},
year = {2025}
}
Comments
arXiv admin note: text overlap with arXiv:2206.08122