Pattern recognition analysis of anterior cingulate cortex blood flow to classify depression polarity
J R C Almeida1, J Mourao-Miranda, H J Aizenstein
1J. R. C. Almeida, MD, PhD, Department of Psychiatry, University of Pittsburgh School of Medicine; J. Mourao-Miranda, PhD, Department of Computer Science, University College London; H. J. Aizenstein, MD, PhD, A. Versace, MD, Department of Psychiatry, University of Pittsburgh School of Medicine; F. A. Kozel, MD, MSCR, Department of Psychiatry and Neuroscience, University of South Florida, Tampa; H. Lu, PhD, Department of Psychiatry, University of Texas Southwestern, Dallas; A. Marquand, PhD, Department of Clinical Neuroscience, Institute of Psychiatry, King's College London; E. J. LaBarbara, BS, Department of Psychiatry, University of Pittsburgh School of Medicine, USA; M. Brammer, PhD, Department of Biostatistics, Institute of Psychiatry, King's College London; M. Trivedi, MD, Department of Psychiatry, University of Texas Southwestern, Dallas; D. J. Kupfer, MD, Department of Psychiatry, University of Pittsburgh School of Medicine; M. L. Phillips, MB, Bchir, MD (Cantab), MRCPsych, Department of Psychiatry, University of Pittsburgh School of Medicine and Department of Psychological Medicine and Clinical Neurosciences, Cardiff University, UK.
None:
Differentiating bipolar from recurrent unipolar depression is a major clinical challenge. In 18 healthy females and 36 females in a depressive episode--18 with bipolar disorder type I, 18 with recurrent unipolar depression--we applied pattern recognition analysis using subdivisions of anterior cingulate cortex (ACC) blood flow at rest, measured with arterial spin labelling. Subgenual ACC blood flow classified unipolar v. bipolar depression with 81% accuracy (83% sensitivity, 78% specificity).
