Machine study of brain scans puts forth new ideas to schizophrenia

The American Psychiatric Association (APA) dismissed a number of different diagnostic classifications for schizophreniain 2013 citing “limited diagnostic stability, low reliability, and poor validity” – the classifications were then based on observations of behavior and symptoms.

Now, new research led by scientists from the University of Pennsylvania (UPenn) using a machine learning method called HYDRA (Heterogeneity Through Discriminate Analysis)reveals significant differences in gray matter volumes and clearly distinguishes two distinct types of schizophrenia – it busts the misconception that all schizophrenia patients have the same brain structure, despite the heterogeneous nature of the disease and notable variations in symptoms and treatment responses from patient to patient.

“Numerous other studies have shown that people with schizophrenia have significantly smaller volumes of brain tissue than healthy controls,” explains UPenn Professor of radiology, Christos Davatzikos. “However, for at least a third of patients we looked at, their brains were almost completely normal.”

As HYDRA scanned over 300 MRI brain scans from schizophrenia patients spanning three continents, 60% of the subjects were found to have decreased gray matter, as predicted, but the remaining 40% displayed virtually normal brains defying common neuroanatomical notions of schizophrenia.

The distinguishing factor between the two schizophrenia subtypes is yet unclear, but this discovery could shape the future of personalised treatment for patients with schizophrenia. Treatments for schizophrenia have differing outcomes in particular people, so it often becomes “a matter of trial and error.”

Category: Education, Features