pubmed.ncbi.nlm.nih.gov

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

  • ️Sat Jan 01 2022

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

Manxi Xu et al. Transl Psychiatry. 2022.

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.

© 2022. The Author(s).

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1. Spatial maps of 23 independent components sorted into eight intrinsic networks overlaid on the Montreal Neurological Institute (MNI) template.

Color bar represents the independent component (IC). AUD auditory network, VIS visual network, SMN somatomotor network, DAN dorsal attention network, VAN ventral attention network, LN limbic network, FPN frontoparietal network, DMN default mode network, L left, R right, STG superior temporal gyrus, Fusiform, fusiform gyrus, Cuneus cuneus cortex, PCC posterior cingulate cortex, Calcarine calcarine cortex, Occipital_Mid middle occipital gyrus, Postcentral postcentral gyrus, Precentral precentral gyrus, SMA supplementary motor area, AG angular gyrus, MFG middle frontal gyrus, Parietal_Sup superior parietal gyrus, PCUN precuneus, IFG inferior frontal gyrus, MiFG middle inferior frontal gyrus, lnsula insular cortex ACC anterior cingulate cortex, SFGmed superior frontal gyrus, medial, SFG superior frontal gyrus, IPL inferior parietal lobule, MTG middle temporal gyrus, mPFC medial prefrontal cortex.

Fig. 2
Fig. 2. State plots and the number of members in each group with each state.

Whole-brain cross-correlation matrices of states 1–6 are shown. The number of participants who entered each state is indicated above the state plots. SA suicide attempter, SI suicide ideation, NS neither SA nor SI, HCs healthy controls, AUD auditory network, VIS visual network, SMN somatomotor network, DAN dorsal attention network, VAN ventral attention network, LN limbic network, FPN frontoparietal network, DMN default mode network.

Fig. 3
Fig. 3. Functional network connectivity differences between patients with major depressive disorder (MDD) and HCs in state 2, state 4, and state 5 and between SAs and HCs in state 4.

Significance was corrected using the false-discovery rate (FDR) over the total of 253 (23 × 22/2) dFNC values in each state. The circles indicate significant (P < 0.05, FDR-corrected) t tests. A wider line means a large group difference. Red lines represent increased connectivity, while blue lines represent decreased connectivity between two groups. AUD auditory network, VIS visual network, SMN, somatomotor network, DAN dorsal attention network, VAN ventral attention network, LN limbic network, FPN frontoparietal network, DMN default mode network, L left, R right, STG superior temporal gyrus, Fusiform fusiform gyrus, Cuneus, cuneus cortex, PCC posterior cingulate cortex, Calcarine calcarine cortex, Occipital_Mid middle occipital gyrus, Postcentral postcentral gyrus; Precentral precentral gyrus, SMA supplementary motor area, AG angular gyrus, MFG middle frontal gyrus, Parietal_Sup superior parietal gyrus, PCUN precuneus, IFG inferior frontal gyrus, MiFG middle inferior frontal gyrus, lnsula insular cortex, ACC anterior cingulate cortex, SFGmed superior frontal gyrus medial, SFG, superior frontal gyrus, IPL inferior parietal lobule, MTG middle temporal gyrus, mPFC medial prefrontal cortex.

Fig. 4
Fig. 4. ROC of the best classifiers between groups.

AUC area under the curve, MDD major depressive disorder, SA suicide attempter, SI suicide ideation, NS individuals had neither SA nor SI, HCs healthy controls.

Fig. 5
Fig. 5. Consensus functional connections in distinguishing patients with major depressive disorder (MDD) from HCs and in distinguishing suicidality among MDD patients.

The brain region of each cluster is represented by a square on the circumference of the big circle. The lines connecting two squares represent the connections between the corresponding two brain regions. The thickness of the line represents the support vector classification weight of the connection. The thicker the line, the larger the weight. Red lines represent positive weight, while blue lines represent negative weight. AUD auditory network, VIS visual network, SMN somatomotor network, DAN dorsal attention network, VAN ventral attention network, LN limbic network, FPN frontoparietal network, DMN default mode network, L left, R right, STG superior temporal gyrus, Fusiform fusiform gyrus, Cuneus cuneus cortex, PCC posterior cingulate cortex, Calcarine calcarine cortex Occipital_Mid middle occipital gyrus, Postcentral postcentral gyrus, Precentral precentral gyrus, SMA supplementary motor area, AG angular gyrus, MFG middle frontal gyrus, Parietal_Sup superior parietal gyrus, PCUN precuneus, IFG inferior frontal gyrus, MiFG middle inferior frontal gyrus, lnsula insular cortex, ACC anterior cingulate cortex, SFGmed superior frontal gyrus medial, SFG superior frontal gyrus, IPL inferior parietal lobule, MTG middle temporal gyrus, mPFC medial prefrontal cortex.

Similar articles

Cited by

References

    1. Woelfer M, Kasties V, Kahlfuss S, Walter M. The role of depressive subtypes within the neuroinflammation hypothesis of Major Depressive Disorder. Neuroscience. 2019;403:93–110. doi: 10.1016/j.neuroscience.2018.03.034. - DOI - PubMed
    1. Mittal VA, Walker EF. Diagnostic and statistical manual of mental disorders. Psychiatry Res. 2011;189:158–9. doi: 10.1016/j.psychres.2011.06.006. - DOI - PMC - PubMed
    1. Nordentoft M, Mortensen PB, Pedersen CB. Absolute risk of suicide after first hospital contact in mental disorder. Arch Gen Psychiatry. 2011;68:1058–64. doi: 10.1001/archgenpsychiatry.2011.113. - DOI - PubMed
    1. Nock MK, Borges G, Bromet EJ, Alonso J, Angermeyer M, Beautrais A, et al. Cross-national prevalence and risk factors for suicidal ideation, plans and attempts. Br J Psychiatry. 2008;192:98–105. doi: 10.1192/bjp.bp.107.040113. - DOI - PMC - PubMed
    1. Scocco P, de Girolamo G, Vilagut G, Alonso J. Prevalence of suicide ideation, plans, and attempts and related risk factors in Italy: results from the European study on the epidemiology of mental disorders-world mental health study. Compr Psychiatry. 2008;49:13–21. doi: 10.1016/j.comppsych.2007.08.004. - DOI - PubMed

Publication types

MeSH terms