Multi-site benchmark classification of major depressive disorder using machine learning on cortical and subcortical measures

Belov, Vladimir, Erwin-Grabner, Tracy, Aghajani, Moji, Aleman, Andre, Amod, Alyssa R., Basgoze, Zeynep, Benedetti, Francesco, Besteher, Bianca, Bülow, Robin, Ching, Christopher R. K., Connolly, Colm G., Cullen, Kathryn, Davey, Christopher G., Dima, Danai, Dols, Annemiek, Evans, Jennifer W., Fu, Cynthia H. Y., Gonul, Ali Saffet, Gotlib, Ian H., Grabe, Hans J., Groenewold, Nynke, Hamilton, J Paul, Harrison, Ben J., Ho, Tiffany C., Mwangi, Benson, Jaworska, Natalia, Jahanshad, Neda, Klimes-Dougan, Bonnie, Koopowitz, Sheri-Michelle, Lancaster, Thomas ORCID: https://orcid.org/0000-0003-1322-2449, Li, Meng, Linden, David E. J. ORCID: https://orcid.org/0000-0002-5638-9292, MacMaster, Frank P., Mehler, David M. A., Melloni, Elisa, Mueller, Bryon A., Ojha, Amar, Oudega, Mardien L., Penninx, Brenda W. J. H., Poletti, Sara, Pomarol-Clotet, Edith, Portella, Maria J., Pozzi, Elena, Reneman, Liesbeth, Sacchet, Matthew D., Sämann, Philipp G., Schrantee, Anouk, Sim, Kang, Soares, Jair C., Stein, Dan J., Thomopoulos, Sophia I., Uyar-Demir, Aslihan, van der Wee, Nic J. A., van der Werff, Steven J. A., Völzke, Henry, Whittle, Sarah, Wittfeld, Katharina, Wright, Margaret J., Wu, Mon-Ju, Yang, Tony T., Zarate, Carlos, Veltman, Dick J., Schmaal, Lianne, Thompson, Paul M. and Goya-Maldonado, Roberto 2024. Multi-site benchmark classification of major depressive disorder using machine learning on cortical and subcortical measures. Scientific Reports 14 (1) , 1084. 10.1038/s41598-023-47934-8

PDF - Published Version Download (2MB)

Abstract

Machine learning (ML) techniques have gained popularity in the neuroimaging field due to their potential for classifying neuropsychiatric disorders. However, the diagnostic predictive power of the existing algorithms has been limited by small sample sizes, lack of representativeness, data leakage, and/or overfitting. Here, we overcome these limitations with the largest multi-site sample size to date (N = 5365) to provide a generalizable ML classification benchmark of major depressive disorder (MDD) using shallow linear and non-linear models. Leveraging brain measures from standardized ENIGMA analysis pipelines in FreeSurfer, we were able to classify MDD versus healthy controls (HC) with a balanced accuracy of around 62%. But after harmonizing the data, e.g., using ComBat, the balanced accuracy dropped to approximately 52%. Accuracy results close to random chance levels were also observed in stratified groups according to age of onset, antidepressant use, number of episodes and sex. Future studies incorporating higher dimensional brain imaging/phenotype features, and/or using more advanced machine and deep learning methods may yield more encouraging prospects.

Item Type:	Article
Date Type:	Publication
Status:	Published
Schools:	Medicine
Additional Information:	License information from Publisher: LICENSE 1: URL: http://creativecommons.org/licenses/by/4.0/, Type: open-access
Publisher:	Nature Research
ISSN:	2045-2322
Date of First Compliant Deposit:	12 January 2024
Date of Acceptance:	19 November 2023
Last Modified:	23 Feb 2024 16:15
URI:	https://orca.cardiff.ac.uk/id/eprint/165426

Actions (repository staff only)

Edit Item