Imaging data ‘clinically useful’ in bipolar diagnosis
By Eleanor McDermid, Senior medwireNews Reporter
18 June 2013
Psychol Med 2013; Advance online publication

medwireNews: Researchers have shown that a pattern recognition technique applied to brain imaging can help to diagnose bipolar disorder.

Rather than simply identifying differences between patients and controls, the team used a type of multivariate pattern recognition algorithm - Gaussian process classifiers (GPCs) - to determine the likelihood for bipolar disorder in individual patients, "thus bridging the gap between neuroscience and clinical practice."

Sophia Frangou (Icahn School of Medicine at Mount Sinai, New York, USA) and co-workers used structural magnetic resonance imaging (sMRI) data for 26 patients with bipolar I disorder and 26 mentally healthy controls matched for age, gender, and IQ. They used some of these matched pairs to "train" the GPCs to distinguish between patients and controls and others to test the predictive ability of the trained GPCs.

"We focused on sMRI, rather than other neuroimaging techniques, as it is widely available, safe and has an established role in the diagnosis and management of brain disorders," the researchers write in Psychological Medicine. "Thus, a diagnostic aid based on sMRI data could be easily incorporated into routine clinical practice and is likely to have high patient acceptability."

The GPCs distinguished between patients and controls with 77% sensitivity and 69% specificity based on gray matter images and 69% sensitivity and 69% specificity for white matter images. In a second cohort of 14 patients and 14 matched controls, the corresponding values were 64% and 99% for gray matter images and 71% and 86% for white matter images. Patients in the second cohort were not taking any medications, bar anticonvulsants, whereas those in the first cohort were taking a variety of medications, including mood stabilizers.

The predictive accuracy of the GPCs was relative to the assessment of qualified psychiatrists using the Structured Clinical Interview for DSM-IV for Axis I Disorders. But Fangou et al say: "A more appropriate comparison would be with behavior-based case-finding instruments, whose sensitivity and specificity are about 70%, such as the Mood Disorder Questionnaire.

"However, even more important is the comparison of our results to 'real world' clinical assessments where [bipolar disorder] is either missed or misdiagnosed resulting in nearly a third of patients having to wait for approximately 10 years before they receive an accurate diagnosis."

They say that given the demonstrable benefits, including cost savings, associated with prompt treatment, early diagnosis of bipolar disorder is "currently the single most important unmet need in enhancing clinical and functional outcomes."

medwireNews ( is an independent clinical news service provided by Springer Healthcare Limited. © Springer Healthcare Ltd; 2013

Free full text

Friendly links

Follow me on Twitter