APSIPA Transactions on Signal and Information Processing > Vol 11 > Issue 1

Maximum Credibility Voting (MCV) – An Integrative Approach for Accurate Diagnosis of Major Depressive Disorder from Clinically Readily Available Data

Yu Shimizu, Okinawa Institute of Science and Technology, Japan, yu.shimizu@aikomi.co.jp , Junichiro Yoshimoto, Nara Institute of Science and Technology, Graduate School of Information Science, Japan, Masahiro Takamura, Hiroshima University, Japan, Go Okada, Hiroshima University, Tomoya Matsumoto, Hiroshima University, Manabu Fuchikami, Hiroshima University, Satoshi Okada, Hiroshima University, Shigeru Morinobu, Hiroshima University, Yasumasa Okamoto, Hiroshima University, Shigeto Yamawaki, Hiroshima University, Kenji Doya, Okinawa Institute of Science and Technology
 
Suggested Citation
Yu Shimizu, Junichiro Yoshimoto, Masahiro Takamura, Go Okada, Tomoya Matsumoto, Manabu Fuchikami, Satoshi Okada, Shigeru Morinobu, Yasumasa Okamoto, Shigeto Yamawaki and Kenji Doya (2022), "Maximum Credibility Voting (MCV) – An Integrative Approach for Accurate Diagnosis of Major Depressive Disorder from Clinically Readily Available Data", APSIPA Transactions on Signal and Information Processing: Vol. 11: No. 1, e14. http://dx.doi.org/10.1561/116.00000042

Publication Date: 23 May 2022
© 2022 Y. Shimizu et al.
 
Subjects
 
Keywords
Biomarkersmajor depressionmachine learningmultimodal
 

Share

Open Access

This is published under the terms of CC BY-NC.

Downloaded: 55 times

In this article:
Introduction 
Methods 
Results 
Discussion 
Conclusion 
References 

Abstract

Diagnosis of Major Depressive Disorder (MDD) is currently a lengthy procedure due to the low diagnostic accuracy of clinically readily available biomarkers. We integrate predictions from multiple datasets based on a credibility parameter defined on the probabilistic distributions of the respective models. We demonstrate by means of structural and resting-state functional magnetic resonance imaging and blood markers obtained from 62 treatment naive MDD patients (age 40.63 ± 9.28, 36 female, HRSD 20.03 ± 4.94) and 66 controls without mental disease history (age 35.52 ± 12.91, 30 female), that our method called Maximum Credibility Voting (MCV) significantly increases diagnostic accuracy from about 65% average classification accuracy of individual biomarker models) to 80% (accuracy after integration of the models). Classification results from different combinations of the available datasets validate the method’s stability with respect to redundant or contradictory predictions. By definition, MCV is applicable to any desired data and compatible with missing values, ensuring continued improvement of diagnostic accuracy and patient comfort as new data acquisition methods and markers emerge.

DOI:10.1561/116.00000042