Journal Article

Prediction of suicidality in bipolar disorder using variability of intrinsic brain activity and machine learning

Wang, Huan, Zhu, Rongxin, Yao, Zhijian, Chattun, Mohammad Ridwan, Wang, Xinyi, Zhang, Siqi, Tian, Shui, Chen, Zhilu, Shao, Junneng, Lu, Qing

2023

Overview

Bipolar disorder (BD) is associated with marked suicidal susceptibility, particularly during a major depressive episode. However, the evaluation of suicidal risk remains challenging since it relies mainly on self‐reported information from patients. Hence, it is necessary to complement neuroimaging features with advanced machine learning techniques in order to predict suicidal behavior in BD patients. In this study, a total of 288 participants, including 75 BD suicide attempters, 101 BD nonattempters and 112 healthy controls, underwent a resting‐state functional magnetic resonance imaging (rs‐fMRI). Intrinsic brain activity was measured by amplitude of low‐frequency fluctuation (ALFF). We trained and tested a two‐level k‐nearest neighbors (k‐NN) model based on resting‐state variability of ALFF with fivefold cross‐validation. BD suicide attempters had increased dynamic ALFF values in the right anterior cingulate cortex, left thalamus and right precuneus. Compared to other machine learning methods, our proposed framework had a promising performance with 83.52% accuracy, 78.75% sensitivity and 87.50% specificity. The trained models could also replicate and validate the results in an independent cohort with 72.72% accuracy. These findings based on a relatively large data set, provide a promising way of combining fMRI data with machine learning technique to reliably predict suicide attempt at an individual level in bipolar depression. Overall, this work might enhance our understanding of the neurobiology of suicidal behavior by detecting clinically defined disruptions in the dynamics of instinct brain activity. We proposed a 2‐levels (voxel‐wise and cluster level) k‐nearest neighbors framework on neurobiomarkers in bipolar depression for early identifying suicide attempters and predicting their suicide risk. The sample size of our study and the validation of an independent sample demonstrate the generalizable classification performance. The most discriminative regions contributing to models were correlated with suicide risk by clinical measurement, further suggesting the neurobiological and clinical underpinnings of our framework.

• Publisher: John Wiley & Sons, Inc
• Subject: Accuracy / amplitude of low‐frequency fluctuation / Biomarkers / Bipolar Disorder / Brain / Brain mapping / Clinical medicine / Cortex (cingulate) / Cortex (parietal) / Depressive Disorder, Major / Functional magnetic resonance imaging / Gyrus Cinguli / Humans / Learning algorithms / Machine learning / Magnetic resonance imaging / Medical imaging / Mental depression / Mental health / Nervous system / Neuroimaging / Psychiatrists / resting‐state functional MRI / Suicidal behavior / Suicidal Ideation / Suicide / suicide attempt / Suicides & suicide attempts / Thalamus