S-Space College of Natural Sciences (자연과학대학) Brain and Cognitive Sciences (뇌인지과학과) Theses (Ph.D. / Sc.D._뇌인지과학과)
Predicting Psychosis Onset in Clinical-High Risk Individuals Using Functional Network Connectivity Biomarkers
기능적 뇌네트워크 간 연결성의 차이를 통한 임상적 고위험군에서 정신증 발병 예측
- 자연과학대학 뇌인지과학과
- Issue Date
- 서울대학교 대학원
- Psychosis; functional MRI; Independent component analysis; Schizophrenia; Clinical-high risk for psychosis; Functional network connectivity
- 학위논문 (박사)-- 서울대학교 대학원 자연과학대학 뇌인지과학과, 2017. 8. 권준수.
- Among individuals at clinical high risk for psychosis (CHR) who show prodromal symptoms of psychosis, some progress to full-blown psychosis. There have been attempts to find markers predicting the onset of psychosis, and brain structures related to onset of psychosis have been reported. However, no studies have examined wide-range interactions at the functional network level that can adequately account for the schizophrenia, a dysconnectivity disorder.
To discover predicting markers for psychosis, I conducted a longitudinal study for a follow-up period of a minimum of 12 months. At the baseline, the resting-state functional magnetic resonance imaging was acquired from individuals at CHR (N = 69), individuals with first-episode psychosis (FEP) (N = 35), and healthy controls (HC) (N = 70). Eight psychosis-related functional networks were extracted, and interactions between paired functional networks were measured, resulting in estimations of 28 possible combinations. After the group comparison, correlation analyses between the altered network interactions and symptom severity were conducted to reveal clinical associations.
Seven of 69 (10%) individuals at CHR proceeded to full-blown psychosis (CHR-C). There were no significant difference in age, gender, and handedness among FEP, CHR-C, CHR-NC, and HC. Of the 28 combinations, there were significant group differences in four functional network connectivity. The FEP group showed the
most severe degree of decrement in functional network connectivity compared to HC. Among all four significantly different functional network interactions, CHR-C showed no significant difference from FEP, while the nonconverters (CHR-NC) had significantly higher functional network connectivity compared to FEP. Among the altered combinations, the interaction between the anterior default mode network and salience network of the FEP group was associated with the overall negative psychotic symptom severity.
This is the first study to suggest that large network interactions can serve as potential markers of the psychosis onset by showing that the CHR-C is similar to FEP, while CHR-NC is comparable to HC. The degree of functional network connectivity in CHR may have prognostic implications regarding the risk of conversion to full-blown psychosis.