Predicting treatment response in psychosis using fMRI: A comprehensive review.
Authors
Affiliations (5)
Affiliations (5)
- Department of Health Sciences, University of Milan, Milan, Italy; Department of Mental Health and Addiction, ASST Santi Paolo e Carlo, Milan, Italy.
- Department of Electronics, Information and Bioengineering, Politecnico di Milano, Milan, Italy; Department of Neurosciences and Mental Health, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy.
- Department of Electronics, Information and Bioengineering, Politecnico di Milano, Milan, Italy.
- Department of Neurosciences and Mental Health, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy; Department of Pathophysiology and Transplantation, University of Milan, Milan, Italy.
- Department of Neurosciences and Mental Health, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy. Electronic address: [email protected].
Abstract
In recent years, the use of functional Magnetic Resonance Imaging (fMRI) methods to predict treatment response in schizophrenia (SCZ) through statistical and machine learning (ML) algorithms has increased. We conducted a comprehensive literature review to assess the role of various fMRI measures in predicting pharmacological treatment response in psychosis. Literature available on PubMed from January 1990 to December 2023 was reviewed, and 21 fMRI studies were included. The results suggest that many studies have employed ML techniques, which may enhance the accuracy of treatment outcome predictions. Additionally, several studies utilizing resting-state fMRI have identified potential associations of functional connectivity patterns across multiple large-scale networks, including the default mode network (DMN), the salience network (SN), the central executive network (CEN), and sensory-motor circuits. These findings suggest that altered connectivity within and between these networks may be relevant for personalized treatment strategies in patients with psychosis, although further investigation is needed to confirm their predictive value. Future research should focus on developing robust and generalizable models to more reliably optimize treatment outcomes in psychosis.