Contents lists available at ScienceDirect Asian Journal of Psychiatry journal homepage: www.elsevier.com/locate/ajp A proof of concept machine learning analysis using multimodal neuroimaging and neurocognitive measures as predictive biomarker in bipolar disorder Rashmin Achalia a,1 , Anannya Sinha b,1 , Arpitha Jacob b , Garimaa Achalia c , Varsha Kaginalkar a , Ganesan Venkatasubramanian b , Naren P. Rao b, * a Government Medical College, Aurangabad, India b National Institute of Mental Health and Neurosciences, Bangalore, India c Achalia Neuropsychiatry Hospital, Aurangabad, Maharashtra, India ARTICLE INFO Keywords: Bipolar disorder Support vector machine Machine learning MRI DTI Neurocognitive test ABSTRACT Background: Concomitant use of complementary, multimodal imaging measures and neurocognitive measures is reported to have higher accuracy as a biomarker in Alzheimer’s dementia. However, such an approach has not been examined to differentiate healthy individuals from Bipolar disorder. In this study, we examined the utility of support vector machine (SVM) technique to differentiate bipolar disorder patients and healthy using struc- tural, functional and diffusion tensor images of brain and neurocognitive measures. Methods: 30 patients with Bipolar disorder-I and 30 age, sex matched individuals participated in the study. Structural MRI, resting state functional MRI and diffusion tensor images were obtained using a 1.5 T scanner. All participants were administered neuropsychological tests to measure executive functions. SVM, a supervised machine learning technique was applied to differentiate patients and healthy individuals with k-fold cross va- lidation over 10 trials. Results: The composite marker consisting of both neuroimaging and neuropsychological measures, had an ac- curacy of 87.60 %, sensitivity of 82.3 % and specificity of 92.7 %. The performance of composite marker was better compared to that of individual markers on classificatory. Conclusions: We were able to achieve a high accuracy for machine learning technique in distinguishing BD from HV using a combination of multimodal neuroimaging and neurocognitive measures. Findings of this proof of concept study, if replicated in larger samples, could have potential clinical applications. 1. Introduction Bipolar disorder (BD) affects 1.5–3 % of the general population and is associated with considerable morbidity, poor occupational func- tioning, and mortality (Merikangas et al., 2007). In contemporary practice, BD is diagnosed based on clinical history, observation, and examination by a psychiatrist; this process carries the risk of subjective bias. The need for an objective biomarker which can differentiate BD patients from healthy individuals and aid in clinical practice is in- creasingly being recognised (Frey et al., 2013; Teixeira et al., 2016). Several candidate measures have been identified as promising bio- markers across three major areas namely neuroimaging, peripheral blood markers, and genetics (Frey et al., 2013). Structural and functional brain abnormalities are typically seen in BD. Loss of grey matter volume and aberrant activity in the subcortical, anterior tem- poral, and ventral prefrontal regions in response to emotional stimuli have been consistently reported in several studies (Alústiza et al., 2017; Chen et al., 2011; Dong et al., 2018; Ganzola and Duchesne, 2017; Pezzoli et al., 2018; Sun et al., 2018) A few studies have examined the utility of machine learning techniques to differentiate BD from healthy individuals using neuroimaging measures as biomarkers (Librenza- Garcia et al., 2017). A majority of such studies have employed Support Vector Machines (SVM) to differentiate BD from healthy individuals or from other disorders like unipolar depression or schizophrenia (Librenza-Garcia et al., 2017). The accuracy levels reported in these studies are mostly in the range of 60 %–80 % (Chen et al., 2014; Fung https://doi.org/10.1016/j.ajp.2020.101984 Received 2 July 2019; Received in revised form 18 February 2020; Accepted 24 February 2020 ⁎ Corresponding author at: Department of Psychiatry, National Institute of Mental Health and Neurosciences, Bangalore, India. E-mail address: docnaren@gmail.com (N.P. Rao). 1 Both authors contributed equally. Asian Journal of Psychiatry 50 (2020) 101984 1876-2018/ © 2020 Elsevier B.V. All rights reserved. T