Ning Mao1, Kaili Che1, Haizhu Xie1, Yuna Li1, Qinglin Wang1, Meijie Liu1, Zhongyi Wang1, Fan Lin1, Heng Ma2, Zhizheng Zhuo3. 1. Department of Radiology, Yantai Yuhuangding Hospital, Qingdao University, Yantai, Shandong, 264000, P. R. China. 2. Department of Radiology, Yantai Yuhuangding Hospital, Qingdao University, Yantai, Shandong, 264000, P. R. China. Electronic address: mahengyhd@163.com. 3. Department of Radiology, Beijing Tiantan Hospital, Beijing, 100044, P. R. China. Electronic address: 18612161210@163.com.
Abstract
BACKGROUND: Postpartum depression (PPD) is a common mental disorder among women. However, the brain information flow alteration in patients with PPD remains unclear. This study investigated the brain information flow characteristics of patients with PPD and their value for clinical evaluation by using support vector regression (SVR). METHODS: Structural and resting-state functional magnetic resonance imaging data were acquired from 21 patients with PPD and 23 age-, educational level-, body mass index-, and menstruation-matched healthy controls. The preferred information flow direction between local brain regions and the preferred information flow direction index within local brain regions based on non-parametric multiplicative regression granger causality analysis were calculated to determine the global and local brain functional characteristics of the patients with PPD. Pearson's correlation analyses were performed to evaluate the relationship of the information flow characteristics with clinical scales. A predictive model for the mental state of the patients with PPD was established using SVR based on information flow characteristics. RESULTS: The information flow patterns in the amygdala, cingulum gyrus, insula, hippocampus, frontal lobe, parietal lobe, and occipital lobe changed significantly in the patients with PPD. The preferred information flow direction between the amygdala and the temporal and frontal lobes significantly correlated with clinical scales. Prediction analysis shows that the information flow patterns can be used to assess depression in patients with PPD. LIMITATION: This exploratory study has a small sample size with no longitudinal research. CONCLUSION: The change in information flow pattern in the amygdala may play an important role in the neuropathological mechanism of PPD and may provide promising markers for clinical evaluation.
BACKGROUND: Postpartum depression (PPD) is a common mental disorder among women. However, the brain information flow alteration in patients with PPD remains unclear. This study investigated the brain information flow characteristics of patients with PPD and their value for clinical evaluation by using support vector regression (SVR). METHODS: Structural and resting-state functional magnetic resonance imaging data were acquired from 21 patients with PPD and 23 age-, educational level-, body mass index-, and menstruation-matched healthy controls. The preferred information flow direction between local brain regions and the preferred information flow direction index within local brain regions based on non-parametric multiplicative regression granger causality analysis were calculated to determine the global and local brain functional characteristics of the patients with PPD. Pearson's correlation analyses were performed to evaluate the relationship of the information flow characteristics with clinical scales. A predictive model for the mental state of the patients with PPD was established using SVR based on information flow characteristics. RESULTS: The information flow patterns in the amygdala, cingulum gyrus, insula, hippocampus, frontal lobe, parietal lobe, and occipital lobe changed significantly in the patients with PPD. The preferred information flow direction between the amygdala and the temporal and frontal lobes significantly correlated with clinical scales. Prediction analysis shows that the information flow patterns can be used to assess depression in patients with PPD. LIMITATION: This exploratory study has a small sample size with no longitudinal research. CONCLUSION: The change in information flow pattern in the amygdala may play an important role in the neuropathological mechanism of PPD and may provide promising markers for clinical evaluation.