Share on Facebook
Share on Twitter
Share on LinkedIn
Share by Email
Print
Brain volume changes in certain areas detected via magnetic resonance imaging (MRI) can predict functioning in patients with bipolar disorder, according to results published in the Journal of Psychiatry Research.
Using a machine learning model, the findings showed that change in left superior frontal cortex, left rostral medial frontal cortex, right white matter total volume, and right lateral ventricle volume can predict reduced functioning in patients with bipolar disorder.
The study included participants with bipolar I disorder (n=35) and healthy controls (n=59). Participants underwent clinical assessments, functioning assessments (Functioning Assessment Short Test [FAST]), and structural MRI. The researchers used machine learning analysis to identify possible candidates for regional brain volume changes that could predict functioning status.
Participants with bipolar disorder and healthy controls did not differ in age, education, or marital status, but there were significant differences in gender, body mass index, FAST score, and employment status.
The researchers observed a significant correlation between observed and predicted FAST scores in participants with bipolar disorder (r=.588, P=.0003), but this was not observed with controls.
The machine learning model determined that the left superior frontal cortex (r=−.66, P <.05), left rostral medial frontal cortex (r=−.66, P<.05), right white matter total volume (r=−.44, P<.05), and right lateral ventricle volume (r=.27, P <.05) could predict FAST scores.
“FAST score was previously reported [to be] not only sensitive to detect[ing] functioning impairment, but also accurate to distinguish early from late stages of [bipolar disorder]. Therefore, it could conceivably integrate a clinical staging model of the disease as the clinical outcome of the neuroprogression of the disorder,” the researchers wrote.
Reference
Sartori JM, Reckziegel R, Passos IC, et al. Volumetric brain resonance imaging predicts functioning in bipolar disorder: a machine learning approach. J Psychiatr Res. 2018; 103:237-243.
