Multimodal Neural Evidence on the Corticostriatal Underpinning of Suicidality in Late-Life Depression

Machine learning and the prediction of suicide in psychiatric populations: a systematic review

Machine learning (ML) has emerged as a promising tool to enhance suicidal prediction. However, as many large-sample studies mixed psychiatric and non-psychiatric populations, a formal psychiatric diagnosis emerged as a strong predictor of suicidal risk, overshadowing more subtle risk factors specific to distinct populations. To overcome this limitation, we conducted a systematic review of ML studies evaluating suicidal behaviors exclusively in psychiatric clinical populations. A systematic literature search was performed from inception through November 17, 2022 on PubMed, EMBASE, and Scopus following the PRISMA guidelines. Original research using ML techniques to assess the risk of suicide or predict suicide attempts in the psychiatric population were included. An assessment for bias risk was performed using the transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD) guidelines. About 1032 studies were retrieved, and 81 satisfied the inclusion criteria and were included for qualitative synthesis. Clinical and demographic features were the most frequently employed and random forest, support vector machine, and convolutional neural network performed better in terms of accuracy than other algorithms when directly compared. Despite heterogeneity in procedures, most studies reported an accuracy of 70% or greater based on features such as previous attempts, severity of the disorder, and pharmacological treatments. Although the evidence reported is promising, ML algorithms for suicidal prediction still present limitations, including the lack of neurobiological and imaging data and the lack of external validation samples. Overcoming these issues may lead to the development of models to adopt in clinical practice. Further research is warranted to boost a field that holds the potential to critically impact suicide mortality.

Grey matter abnormalities in major depressive disorder patients with suicide attempts: A systematic review of age-specific differences

Background

Previous studies have reported alterations in brain structure in major depressive disorder (MDD) patients with suicide attempts. However, age-related changes in suicidal MDD patients remain unclear.

Methods

We performed a systematic review following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Embase, PubMed, and Web of Science were searched to identify relevant studies from inception to January 2023. All voxel-based and surface-based morphometry studies comparing suicidal MDD patients to MDD or healthy controls were included. Studies were then grouped by age range (old, middle-age, adolescent) and the commonalities and age-related structural brain alterations were summarized. The included studies were evaluated using the Newcastle-Ottawa Scale (NOS).

Results

A total of 17 studies met the inclusion criteria, including 3 of late-life depression (LLD) patients, 11 of middle-aged depression (MAD) patients, and 3 of adolescent depression (AOD) patients. The majority of studies had moderate to high NOS scores, indicating good quality. Patients in all three age groups exhibited extensive alterations in the lateral, medial, and orbital regions of the frontal lobes. Furthermore, suicidal MAD patients showed a specific decrease in the gray matter volume of the dorsolateral prefrontal cortex compared to suicidal LLD patients. Cortical thickness and left angular gyrus volume were decreased in suicidal MAD and suicidal LLD patients, but increased in suicidal AOD patients.

Conclusion

This systematic review summarizes structural brain changes in suicidal MDD patients at three age groups: elderly, middle-aged, and adolescent. These findings help elucidate the common circuitry of MDD related to suicide over the lifespan and highlight unique circuitry associated with different ages. These findings may help predict the risk of suicide in MDD patients at different ages.

Predicting suicidality in late-life depression by 3D convolutional neural network and cross-sample entropy analysis of resting-state fMRI

BACKGROUND

Predicting suicide is a pressing issue among older adults; however, predicting its risk is difficult. Capitalizing on the recent development of machine learning, considerable progress has been made in predicting complex behavior such as suicide. As depression remained the strongest risk for suicide, we aimed to apply deep learning algorithms to identify suicidality in a group with late-life depression (LLD).

METHODS

We enrolled 83 patients with LLD, 35 of which were non-suicidal and 48 were suicidal, including 26 with only suicidal ideation and 22 with past suicide attempts, for resting-state functional magnetic resonance imaging (MRI). Cross-sample entropy (CSE) analysis was conducted to examine the complexity of MRI signals among brain regions. Three-dimensional (3D) convolutional neural networks (CNNs) were used, and the classification accuracy in each brain region was averaged to predict suicidality after sixfold cross-validation.

RESULTS

We found brain regions with a mean accuracy above 75% to predict suicidality located mostly in default mode, fronto-parietal, and cingulo-opercular resting-state networks. The models with right amygdala and left caudate provided the most reliable accuracy in all cross-validation folds, indicating their neurobiological importance in late-life suicide.

CONCLUSION

Combining CSE analysis and the 3D CNN, several brain regions were found to be associated with suicidality.

Electric field distribution predicts efficacy of accelerated intermittent theta burst stimulation for late-life depression

Introduction

Repetitive transcranial magnetic stimulation (rTMS) is a promising intervention for late-life depression (LLD) but may have lower rates of response and remission owing to age-related brain changes. In particular, rTMS induced electric field strength may be attenuated by cortical atrophy in the prefrontal cortex. To identify clinical characteristics and treatment parameters associated with response, we undertook a pilot study of accelerated fMRI-guided intermittent theta burst stimulation (iTBS) to the right dorsolateral prefrontal cortex in 25 adults aged 50 or greater diagnosed with LLD and qualifying to receive clinical rTMS.

Methods

Participants underwent baseline behavioral assessment, cognitive testing, and structural and functional MRI to generate individualized targets and perform electric field modeling. Forty-five sessions of iTBS were delivered over 9 days (1800 pulses per session, 50-min inter-session interval). Assessments and testing were repeated after 15 sessions (Visit 2) and 45 sessions (Visit 3). Primary outcome measure was the change in depressive symptoms on the Inventory of Depressive Symptomatology-30-Clinician (IDS-C-30) from Visit 1 to Visit 3.

Results

Overall there was a significant improvement in IDS score with the treatment (Visit 1: 38.6; Visit 2: 31.0; Visit 3: 21.3; mean improvement 45.5%) with 13/25 (52%) achieving response and 5/25 (20%) achieving remission (IDS-C-30 < 12). Electric field strength and antidepressant effect were positively correlated in a subregion of the ventrolateral prefrontal cortex (VLPFC) (Brodmann area 47) and negatively correlated in the posterior dorsolateral prefrontal cortex (DLPFC).

Conclusion

Response and remission rates were lower than in recently published trials of accelerated fMRI-guided iTBS to the left DLPFC. These results suggest that sufficient electric field strength in VLPFC may be a contributor to effective rTMS, and that modeling to optimize electric field strength in this area may improve response and remission rates. Further studies are needed to clarify the relationship of induced electric field strength with antidepressant effects of rTMS for LLD.

Brain morphometric features predict depression symptom phenotypes in late-life depression using a deep learning model

Objectives

Our objective was to use deep learning models to identify underlying brain regions associated with depression symptom phenotypes in late-life depression (LLD).

Participants

Diagnosed with LLD (N = 116) and enrolled in a prospective treatment study.

Design

Cross-sectional.

Measurements

Structural magnetic resonance imaging (sMRI) was used to predict five depression symptom phenotypes from the Hamilton and MADRS depression scales previously derived from factor analysis: (1) Anhedonia, (2) Suicidality, (3) Appetite, (4) Sleep Disturbance, and (5) Anxiety. Our deep learning model was deployed to predict each factor score via learning deep feature representations from 3D sMRI patches in 34 a priori regions-of-interests (ROIs). ROI-level prediction accuracy was used to identify the most discriminative brain regions associated with prediction of factor scores representing each of the five symptom phenotypes.

Results

Factor-level results found significant predictive models for Anxiety and Suicidality factors. ROI-level results suggest the most LLD-associated discriminative regions in predicting all five symptom factors were located in the anterior cingulate and orbital frontal cortex.

Conclusions

We validated the effectiveness of using deep learning approaches on sMRI for predicting depression symptom phenotypes in LLD. We were able to identify deep embedded local morphological differences in symptom phenotypes in the brains of those with LLD, which is promising for symptom-targeted treatment of LLD. Future research with machine learning models integrating multimodal imaging and clinical data can provide additional discriminative information.

[Morphofunctional cerebral changes associated with development of suicidal behavior]

THE AIM OF THE STUDY

Was to identify the cerebral areas, which demonstrate the most significant structural changes and damaged functional activity in patients with suicidal behavior. The original studies, presented in PubMed database, were used to analyze the literature. Additional literature in the form of atlases, review articles and publications, written in related spheres, was used to interpret the results. The study identified the 69 cerebral regions, demonstrating significant changes and the structures with the most significant deviations among them were selected. The regions of cerebral grey matter, in particular basal ganglia (structures of striatum and limbic system), as well as selected regions of cerebral cortex, specifically frontal, insularis, singulate and parietal mostly were included in the list. The decrease in grey matter volume, changes of neuronal and glial density, special patterns of activity and variations of functional association with other cerebral regions are described within mentioned structures. The literature review found that there was a lack of postmortem examinations in suicidal cases. Advanced study of the described structures is required in cases of completed suicide using new research methods.