M Li1,2,3, T Das4,5,6, W Deng1,2,3, Q Wang1,2,3, Y Li1,2,3, L Zhao1,2,3, X Ma1,2,3, Y Wang1,2,3, H Yu1,2,3, X Li1,2,3, Y Meng1,2,3, L Palaniyappan4,5,6, T Li1,2,3. 1. Mental Health Center, West China Hospital of Sichuan University, Chengdu, China. 2. State Key Laboratory of Biotherapy, Psychiatric Laboratory, West China Hospital, Sichuan University, Chengdu, Sichuan, China. 3. West China Brain Research Center, West China Hospital of Sichuan University, Chengdu, China. 4. Robarts Research Institute & The Brain and Mind Institute, University of Western Ontario, London, ON, Canada. 5. Department of Psychiatry, University of Western Ontario, London, ON, Canada. 6. Lawson Health Research Institute, London, ON, Canada.
Abstract
OBJECTIVE: Depression in bipolar disorder (BipD) requires a therapeutic approach that is from treating unipolar major depressive disorder (UniD), but to date, no reliable methods could separate these two disorders. The aim of this study was to establish the clinical validity and utility of a non-invasive functional MRI-based method to classify BipD from UniD. METHOD: The degree of connectivity (degree centrality or DC) of every small unit (voxel) with every other unit of the brain was estimated in 22 patients with BipD and 22 age, gender, and depressive severity-matched patients with UniD and 22 healthy controls. Pattern classification analysis was carried out using a support-vector machine (SVM) approach. RESULTS: Degree centrality pattern from 8-min resting fMRI discriminated BipD from UniD with an accuracy of 86% and diagnostic odds ratio of 9.6. DC was reduced in the left insula and increased in bilateral precuneus in BipD when compared to UniD. In this sample with a high degree of uncertainty (50% prior probability), positive predictive value of the DC test was 79%. CONCLUSION: Degree centrality maps are potential candidate measures to separate bipolar depression from unipolar depression. Test performance reported here requires further pragmatic evaluation in regular clinical practice.
OBJECTIVE:Depression in bipolar disorder (BipD) requires a therapeutic approach that is from treating unipolar major depressive disorder (UniD), but to date, no reliable methods could separate these two disorders. The aim of this study was to establish the clinical validity and utility of a non-invasive functional MRI-based method to classify BipD from UniD. METHOD: The degree of connectivity (degree centrality or DC) of every small unit (voxel) with every other unit of the brain was estimated in 22 patients with BipD and 22 age, gender, and depressive severity-matched patients with UniD and 22 healthy controls. Pattern classification analysis was carried out using a support-vector machine (SVM) approach. RESULTS: Degree centrality pattern from 8-min resting fMRI discriminated BipD from UniD with an accuracy of 86% and diagnostic odds ratio of 9.6. DC was reduced in the left insula and increased in bilateral precuneus in BipD when compared to UniD. In this sample with a high degree of uncertainty (50% prior probability), positive predictive value of the DC test was 79%. CONCLUSION: Degree centrality maps are potential candidate measures to separate bipolar depression from unipolar depression. Test performance reported here requires further pragmatic evaluation in regular clinical practice.
Authors: Leah R Kling; Katie L Bessette; Sophie R DelDonno; Kelly A Ryan; Wayne C Drevets; Melvin G McInnis; Mary L Phillips; Scott A Langenecker Journal: Bipolar Disord Date: 2018-10-07 Impact factor: 6.744