Machine Learning based identification of suicidal risk in patients with schizophrenia using multi-level resting state fMRI features

Abstract

Background: Some studies suggest that as much as 40% of all causes of death in a group of patients with schizophrenia can be attributed to suicides and compared with the general population, patients with schizophrenia have an 8.5-fold greater suicide risk (SR). There is a vital need for accurate and reliable methods to predict the SR among patients with schizophrenia based on biological measures. However, it is unknown whether the suicidal risk in schizophrenia can be related to alterations in spontaneous brain activity, or if the resting-state functional magnetic resonance imaging (rsfMRI) measures can be used alongside machine learning (ML) algorithms in order to identify patients with SR.Methods: Fifty-nine participants including patients with schizophrenia with and without SR as well as age and gender-matched healthy underwent 13 minutes of resting-state functional magnetic resonance imaging. Both static and dynamic indexes of the amplitude of low-frequency fluctuation (ALFF), the fractional amplitude of low-frequency fluctuations (fALFF), regional homogeneity as well as functional connectivity (FC) was calculated and used as an input for five machine learning algorithms: Gradient boosting (GB), LASSO, Logistic Regression (LR), Random Forest and Support Vector Machine. Results: All groups revealed different internetwork functional connectivity in ventral DMN and anterior SN. The best performance was reached for the LASSO applied to FC with an accuracy of 70% and AUROC of 0.76 (p<0.05). Significant classification ability was also reached for GB and LR using fALFF and ALFF measures.Conclusion: Our findings suggest that SR in schizophrenia can be seen on the level of DMN and SN functional connectivity alterations. ML algorithms were able to significantly differentiate SR patients. Our results could be useful in developing neuromarkers of SR in schizophrenia based on non-invasive rsfMRI.

Neuroimaging biomarkers for detecting schizophrenia: A resting-state functional MRI-based radiomics analysis

“…The rs-fMRI values in neuropsychiatric disorders have been confirmed ( Li et al., 2019a ; Qiu et al., 2020 ; Zhao et al., 2020 ; Zhou et al., 2020a ), including SZ ( Bohaterewicz et al., 2020 ; Li et al., 2019a , 2020 ; Xiao et al., 2019 ). Among them, FC is a commonly used indicator ( Bohaterewicz et al., 2020 ; Li et al., 2019a ; Lin et al., 2020 ). However, conventional FC evaluates connectivity patterns between predefined specific brain regions or distinct brain network components by the ICA method.…”

“…Previous studies have mostly used group-level comparison analysis ( Lu et al., 2021 ; Xue et al., 2019 ; Yan et al., 2020 ; Zhao et al., 2018b ) or machine learning ( Bohaterewicz et al., 2020 ; Li et al., 2019a , 2020 ) to evaluate the spontaneous neural activity abnormalities of SZ patients. This study was the first that we are aware of to apply a radiomics-based machine learning method with DC and VMHC metrics to detect SZ potential biomarkers and explore the potential biological mechanism of SZ.…”

“…Previous studies ( Jiang et al., 2020 ; Li et al., 2019a , 2020 ; Shi et al., 2021a ; Wang et al., 2020a ; Xiao et al., 2019 ) proposed that MRI measures could be used as potential biomarkers for SZ. Among them, functional connectivity (FC) is a commonly used metric and it can be static ( Bohaterewicz et al., 2020 ; de Filippis et al., 2019 ; Jiang et al., 2020 ; Li et al., 2020 ; Orban et al., 2018 ; Plis et al., 2018 ) or dynamic ( Luo et al., 2019 ; Sun et al., 2019 ; Weber et al., 2020 ; Zhang et al., 2021 ). However, conventional FC evaluates connectivity patterns between predefined brain regions or distinct brain network components via independent component analysis (ICA).…”