Identification of suicidality in patients with major depressive disorder via dynamic functional network connectivity signatures and machine learning

Abstract

Major depressive disorder (MDD) is a severe brain disease associated with a significant risk of suicide. Identification of suicidality is sometimes life-saving for MDD patients. We aimed to explore the use of dynamic functional network connectivity (dFNC) for suicidality detection in MDD patients. A total of 173 MDD patients, including 48 without suicide risk (NS), 74 with suicide ideation (SI), and 51 having attempted suicide (SA), participated in the present study. Thirty-eight healthy controls were also recruited for comparison. A sliding window approach was used to derive the dFNC, and the K-means clustering method was used to cluster the windowed dFNC. A linear support vector machine was used for classification, and leave-one-out cross-validation was performed for validation. Other machine learning methods were also used for comparison. MDD patients had widespread hypoconnectivity in both the strongly connected states (states 2 and 5) and the weakly connected state (state 4), while the dysfunctional connectivity within the weakly connected state (state 4) was mainly driven by suicidal attempts. Furthermore, dFNC matrices, especially the weakly connected state, could be used to distinguish MDD from healthy controls (area under curve [AUC] = 82), and even to identify suicidality in MDD patients (AUC = 78 for NS vs. SI, AUC = 88 for NS vs. SA, and AUC = 74 for SA vs. SI), with vision-related and default-related inter-network connectivity serving as important features. Thus, the dFNC abnormalities observed in this study might further improve our understanding of the neural substrates of suicidality in MDD patients.

The influence of childhood emotional neglect experience on brain dynamic functional connectivity in young adults

“…The similarity between windowed functional connectivity matrices was calculated using the default distance measures (sqEuclidean distance). It has been proven to be an effective measure for high-dimensional data (Lin et al, 2021 ; Xu et al, 2022 ). Furthermore, we performed a cluster validity analysis to evaluate the optimal number of states based on three criteria (Silhouette, Davies-Bouldin values, and Calinski – Harabasz), and the maximum number of states allowed to estimate was set at 10.…”

Functional brain network dynamics of brooding in depression: insights from real-time fMRI neurofeedback

“…The temporal dynamics of these connections are particularly relevant to MDD. MDD has been associated with more time spent having reduced FC within somatomotor and dorsal attention networks (Javaheripour et al, 2023), less time in integrated states with increased FC between sensory and default-mode networks (Wu et al, 2019), and more time with reduced FC within and between visual, auditory, somatomotor and defaultmode networks (Xu et al, 2022). Similarly, recent mega-and meta-analytic evidence of prominent static FC alterations in MDD implicated hypoconnectivity within and between dorsal attention, somatomotor, parietal and visual networks (Javaheripour et al, 2021;Tse et al, 2023).…”

Predictive Models for Suicide Attempts in Major Depressive Disorder and the Contribution of EPHX2: A Pilot Integrative Machine Learning Study

“…Concentrating on high-risk subgroups, such as individuals with major depressive disorder (MDD), could offer a solution to this issue [4,5]. Notably, MDD is a significant contributor to the disease burden, characterized by an exceptionally high risk of suicide [6]. Early recognition of depressive symptoms is one of the critical facets in suicide prevention and may save the lives of patients with MDD [6,7].…”

“…Notably, MDD is a significant contributor to the disease burden, characterized by an exceptionally high risk of suicide [6]. Early recognition of depressive symptoms is one of the critical facets in suicide prevention and may save the lives of patients with MDD [6,7]. Therefore, focusing on subgroups of patients with MDD to establish a suicidal predictive model can improve our ability in suicide prediction and prevention.…”