Functional connectome-based predictive modeling of suicidal ideation.
Authors
Affiliations (10)
Affiliations (10)
- Baylor College of Medicine, Menninger Department of Psychiatry and Behavioral Sciences, 1977 Butler Boulevard, Houston, TX 77030, USA; Michael E. DeBakey VA Medical Center, 2002 Holcombe Boulevard, Houston, TX 77030, USA; Yale School of Medicine, 333 Cedar St, New Haven, CT 06510, USA.
- Baylor College of Medicine, Menninger Department of Psychiatry and Behavioral Sciences, 1977 Butler Boulevard, Houston, TX 77030, USA; Michael E. DeBakey VA Medical Center, 2002 Holcombe Boulevard, Houston, TX 77030, USA.
- Duke University School of Medicine, Department of Psychiatry, 905 W Main St, Durham, NC 27701, USA.
- Baylor College of Medicine, Menninger Department of Psychiatry and Behavioral Sciences, 1977 Butler Boulevard, Houston, TX 77030, USA; Michael E. DeBakey VA Medical Center, 2002 Holcombe Boulevard, Houston, TX 77030, USA; Yale School of Medicine, 333 Cedar St, New Haven, CT 06510, USA; Baylor College of Medicine, Core for Advanced Magnetic Resonance Imaging (CAMRI), One Baylor Plaza, Suite S104, Houston, TX 77030, USA.
- University of South Carolina, Department of Communication Sciences and Disorders, 1705 College Street, Columbia, SC, USA.
- Baylor College of Medicine, Menninger Department of Psychiatry and Behavioral Sciences, 1977 Butler Boulevard, Houston, TX 77030, USA; University of Cambridge, Department of Psychiatry, Addenbrooke's Hospital, Hills Road, Cambridge CB2 2QQ, UK.
- Baylor College of Medicine, Menninger Department of Psychiatry and Behavioral Sciences, 1977 Butler Boulevard, Houston, TX 77030, USA; Baylor College of Medicine, Department of Neuroscience, Department of Neuroscience One Baylor Plaza, S640, Houston, TX 77030, USA.
- Department of Psychiatry and Behavioral Sciences, Stanford University, 401 Quarry Rd, Palo Alto, CA 94304, USA.
- Baylor College of Medicine, Menninger Department of Psychiatry and Behavioral Sciences, 1977 Butler Boulevard, Houston, TX 77030, USA; Michael E. DeBakey VA Medical Center, 2002 Holcombe Boulevard, Houston, TX 77030, USA; Baylor College of Medicine, Department of Neuroscience, Department of Neuroscience One Baylor Plaza, S640, Houston, TX 77030, USA; Michael E. DeBakey VA Medical Center, Center for Translational Research on Inflammatory Diseases, 2002 Holcombe Boulevard, Houston, TX 77030, USA; The Menninger Clinic, 12301 S Main Street, Houston, TX 77035, USA.
- Baylor College of Medicine, Menninger Department of Psychiatry and Behavioral Sciences, 1977 Butler Boulevard, Houston, TX 77030, USA; Michael E. DeBakey VA Medical Center, 2002 Holcombe Boulevard, Houston, TX 77030, USA; Yale School of Medicine, 333 Cedar St, New Haven, CT 06510, USA; Baylor College of Medicine, Core for Advanced Magnetic Resonance Imaging (CAMRI), One Baylor Plaza, Suite S104, Houston, TX 77030, USA. Electronic address: [email protected].
Abstract
Suicide represents an egregious threat to society despite major advancements in medicine, in part due to limited knowledge of the biological mechanisms of suicidal behavior. We apply a connectome predictive modeling machine learning approach to identify a reproducible brain network associated with suicidal ideation in the hopes of demonstrating possible targets for novel anti-suicidal therapeutics. Patients were recruited from an inpatient facility at The Menninger Clinic, in Houston, Texas (N = 261; 181 with active and specific suicidal ideation) and had a current major depressive episode and recurrent major depressive disorder, underwent resting-state functional magnetic resonance imaging. The participants' ages ranged from 18 to 70 (mean ± SEM = 31.6 ± 0.8 years) and 136 (52 %) were males. Using this approach, we found a robust and reproducible biomarker of suicidal ideation relative to controls without ideation, showing that increased suicidal ideation was associated with greater internal connectivity and reduced internetwork external connectivity in the central executive, default mode, and dorsal salience networks. We also found evidence for higher external connectivity between ventral salience and sensorimotor/visual networks as being associated with increased suicidal ideation. Overall, these observed differences may reflect reduced network integration and higher segregation of connectivity in individuals with increased suicide risk. Our findings provide avenues for future work to test novel drugs targeting these identified neural alterations, for instance drugs that increase network integration.