Receive a weekly summary and discussion of the top papers of the week by leading researchers in the field.

In Nan fang yi ke da xue xue bao = Journal of Southern Medical University

OBJECTIVE : To investigate the changes of functional connectivity (FC) in the suprachiasmatic nucleus (SCN) of patients with bipolar disorder and perform a cluster analysis of patients with bipolar disorder based on FC.

METHODS : The study recruited 138 patients with bipolar disorder (BD) diagnosed according to the 4th edition of Diagnostic and Statistical Manual of Mental Disorders IV (DSM-IV) and 150 healthy control subjects. All the participants underwent resting-state functional magnetic resonance brain scans. DPARSF software was used to generate the FC diagram of the SCN. Based on the FC data, principal components analysis (PCA) and k-means in scikit-learn 0.20.1 were used for cluster analysis of the patients with bipolar disorder.

RESULTS : Compared with the healthy controls, the patients showed enhanced functional connections between the SCN and the paraventricular nucleus and between the SCN and the dorsomedial hypothalamus nucleus. Based on these FC values, the optimal cluster of unsupervised k-means machine learning for bipolar disorder was 2, and the Silhouette coefficient was 0.49.

CONCLUSIONS : Patients with bipolar disorder have changes in the FC of the SCN, and the FC of the rhythm pathway can divide bipolar disorder into two subtypes, suggesting that biological rhythm is one of the potential biomarkers of bipolar disorder.

Liu Manli, Meng Yajing, Wei Wei, Li Tao

2020-Jun-30

biological rhythm, bipolar disorder, functional connectivity, machine learning, suprachiasmatic nucleus