COSTA MESA, Calif., July 6, 2026 /PRNewswire/ -- A groundbreaking new study published in NeuroImage: Reports demonstrates that whole-brain Single Photon Emission Computed Tomography (SPECT) imaging, combined with advanced machine learning techniques, can successfully differentiate individuals with schizophrenia from healthy controls with a high degree of accuracy. The research represents an important step toward developing objective brain-based tools to support psychiatric diagnosis and treatment.

The study, led by researchers from the Tri-Institutional Center for Translational Research in Neuroimaging and Data Science (TReNDS), Amen Clinics, and collaborating institutions, analyzed brain SPECT scans from 213 individuals, including 137 patients diagnosed with schizophrenia and 76 healthy controls. Researchers used a sophisticated brain-network analysis approach known as spatially constrained Independent Component Analysis (sc-ICA), guided by the NeuroMark functional brain network template, to identify patterns of altered brain function associated with schizophrenia.

The resulting brain network features were then evaluated using multiple machine learning algorithms. Surprisingly, traditional Support Vector Machine (SVM) models, which have shown promise in functional MRI studies, were outperformed by logistic regression and random forest classifiers. Logistic regression achieved 87% sensitivity and 68% specificity, while random forest reached 88% sensitivity and 61% specificity in distinguishing schizophrenia patients from healthy subjects.

Researchers found that abnormalities within visual processing and cognitive control networks emerged as some of the most predictive features of schizophrenia. Key brain regions included the middle occipital gyrus, subthalamus, and putamen, reinforcing growing evidence that schizophrenia involves widespread disruptions across interconnected brain systems rather than isolated abnormalities.

"This work demonstrates that brain SPECT imaging contains meaningful network-level information that can be leveraged by machine learning to improve our understanding of schizophrenia," said co-author Dr. Daniel Amen, founder and CEO of Amen Clinics. "These findings further support the concept that psychiatric disorders are brain disorders that can be studied objectively through functional neuroimaging."

The authors believe these findings could eventually contribute to more personalized approaches for assessing risk, monitoring treatment response, and identifying brain circuits associated with symptoms such as hallucinations, delusions, and cognitive impairment. Future studies with larger and more balanced populations are planned to further validate the results and explore multimodal imaging approaches.

About Amen Clinics
Amen Clinics is a nationwide network of brain health centers founded by Daniel G. Amen, MD, a board-certified psychiatrist, physician, researcher, and bestselling author. For more than 35 years, Amen Clinics has pioneered a brain-health approach to mental wellness, helping patients better understand the connection between brain function and emotional, behavioral, cognitive, and mental health challenges. With 11 clinics across the United States the world's largest database of brain SPECT scans related to behavior, Amen Clinics is dedicated to changing the conversation from mental illness to brain health

Media Contact:

Jordyn Dean

Publicist, Amen Clinics

jordyn.dean@amenclinics.com

View original content to download multimedia:https://www.prnewswire.com/news-releases/new-study-reveals-powerful-new-method-can-distinguish-schizophrenia-from-healthy-using-brain-spect-imaging-302817844.html

SOURCE Amen Clinics, Inc.