Predicting Suicide Attempts among Major Depressive Disorder Patients with Structural Neuroimaging: A Machine Learning Approach

Introduction Every year at least one million people die by suicide, with major depressive disorder (MDD) being one of the major causes of suicide deaths. Current suicide risk assessments rely on subjective information, are time consuming, low predictive, and poorly reliable. Thus, finding objective biomarkers of suicidality is crucial to move clinical practice towards a precision psychiatry framework, enhancing suicide risk detection and prevention for MDD. Objectives The present study aimed at applying machine learning (ML) algorithms on both grey matter and white-matter voxel-wise data to discriminate MDD suicide attempters (SA) from non-attempters (nSA). Methods 91 currently depressed MDD patients (24 SA, 67 nSA) underwent a structural MRI session. T1-weighted images and diffusion tensor imaging scans were respectively pre-processed using Computational Atlas Toolbox 12 (CAT12) and spatial tract-based statistics (TBSS) on FSL, to obtain both voxel-based morphometry (VBM) and fractional anisotropy (FA) measures. Three classification models were built, entering whole-brain VBM and FA maps alone into a Support Vector Machine (SVM) and combining both modalities into a Multiple Kernel Learning (MKL) algorithm. All models were trained through a 5-fold nested cross-validation with subsampling to calculate reliable estimates of balanced accuracy, specificity, sensitivity, and area under the receiver operator curve (AUC). Results Models’ performances are summarized in Table 1.Table 1. Models’ performances. Input features Algorithm Specificity Sensitivity Balanced accuracy AUC VBM SVM 55.00% 50.00% 52.50% 0.55 FA SVM 72.00% 54.00% 63.00% 0.62 VBM and FA MKL 68.00% 54.00% 61.00% 0.58 Abbreviations: AUC, area under the receiver operator curve; FA, fractional anisotropy; VBM, voxel-based morphometry. Conclusions Overall, although overcoming the random classification accuracy (i.e., 50%), performances of all models classifying SA and nSA MDD patients were moderate, possibly due to the imbalanced numerosity of classes, with SVM on FA reaching the highest accuracy. Thus, future studies may enlarge the sample and add different features (e.g., functional neuroimaging data) to develop an objective and reliable predictive model to assess and hence prevent suicide risk among MDD patients. Disclosure of Interest None Declared

Introduction: Irish Travellers are an indigenous minority group in Ireland. Health inequalities have been widely reported within the Traveller community, with a shorter life expectancy of 11 years less than the general population. Travellers also have higher mortality rates of 3.5 times higher than the general population in Ireland. Suicide is a serious problem in the Traveller community with a suicide rate of 11% among Travellers: 6 times higher in women and 7 times higher in men compared with their counterparts in the general population.
Objectives: There is a paucity of research into the clinical characteristics of self-harm and suicidality among Irish Travellers despite the elevated suicide rates in this community. This study aims to bridge the knowledge gap in the mental health of Irish Travellers, focusing on the clinical factors associated with self-harm and suicidality in a community sample of Irish Travellers.
Methods: This is a cross-sectional study. Study participants completed self-report and interview-based validated questionnaires that screen for anxiety (General Anxiety Disorder assessment: GAD-7), depression (Patient Health Questionnaire: PHQ-9), and suicidality (Suicide Behaviours Questionnaire-Revised: SBQ-R and Adult Suicidal Ideation Questionnaire: ASIQ). Ethical approval was granted through the Clinical Research Ethics Committee, University College Dublin. Results: Despite an active recruitment campaign, participation rate from Irish Travellers in this study was low, with only five participants completing this study. Three were male. The mean age of the study participants was 39AE14.7 years. All had pre-existing mental health diagnoses, most commonly anxiety disorder. All had at least one previous episode of self-harm and 80% had a positive family history of self-harming behaviour. No participants reported a history of alcohol or substance misuse. Over half of the participants reported severe anxiety and depressive symptoms with median GAD-7 score of 19 and PHQ-9 score of 21 respectively. All participants demonstrated significant risk of suicidal behaviour based on their SBQ-R and ASIQ scores. Conclusions: Despite elevated rates of suicidality and mental illness in this ethnic minority group, Irish Travellers demonstrated lower participation in mental health research, including this study. These recruitment challenges suggest that factors such as stigma, shame and lack of trust may be contributory. These factors may also act as barriers to them accessing mental healthcare when they develop mental health symptoms such as anxiety and depression, associated with increased risk of self-harm and suicidal behaviours. There is a need for better engagement strategies with Travellers to promote awareness into their needs and reduce mental health problems in this population. Introduction: Every year at least one million people die by suicide, with major depressive disorder (MDD) being one of the major causes of suicide deaths. Current suicide risk assessments rely on subjective information, are time consuming, low predictive, and poorly reliable. Thus, finding objective biomarkers of suicidality is crucial to move clinical practice towards a precision psychiatry framework, enhancing suicide risk detection and prevention for MDD.

Disclosure of
Objectives: The present study aimed at applying machine learning (ML) algorithms on both grey matter and white-matter voxel-wise data to discriminate MDD suicide attempters (SA) from nonattempters (nSA). Methods: 91 currently depressed MDD patients (24 SA, 67 nSA) underwent a structural MRI session. T1-weighted images and diffusion tensor imaging scans were respectively pre-processed using Computational Atlas Toolbox 12 (CAT12) and spatial tract-based statistics (TBSS) on FSL, to obtain both voxel-based morphometry (VBM) and fractional anisotropy (FA) measures. Three classification models were built, entering whole-brain VBM and FA maps alone into a Support Vector Machine (SVM) and combining both modalities into a Multiple Kernel Learning (MKL) algorithm. All models were trained through a 5-fold nested cross-validation with subsampling to calculate reliable estimates of balanced accuracy, specificity, sensitivity, and area under the receiver operator curve (AUC).
Results: Models' performances are summarized in Table 1. Conclusions: Overall, although overcoming the random classification accuracy (i.e., 50%), performances of all models classifying SA and nSA MDD patients were moderate, possibly due to the imbalanced numerosity of classes, with SVM on FA reaching the highest accuracy. Thus, future studies may enlarge the sample and add different features (e.g., functional neuroimaging data) to develop an objective and reliable predictive model to assess and hence prevent suicide risk among MDD patients. Introduction: More than 90 percent of patients who attempt suicide have a psychiatric disorder. The diagnosis of schizophrenia is associated with a decrease in life expectancy of about 10 years, with suicide being the most important related factor. Literature suggests that the risk of suicide death in this population has been found to be 10 to 20 times higher than that in the general population Objectives: To present a case report of a patient with a first psychotic episode and suicide attempt focusing on clinical features and risk factors. Methods: Presentation of a clinical case supported by a nonsystematic review of literature containing the key-words "suicide", "Suicidal ideation", "psychosis" and "schizophrenia". Results: This is a case report of a male 28-year-old patient, with no known psychiatric history, admitted to our inpatient service after a suicide attempt by precipitation. In a first evaluation, the patient presented psychotic symptoms consisting of paranoid delusions, auditory hallucinations, tendency to social isolation and the appearance of self-harming ideation in the days prior to the episode. After initiation of antipsychotic medication, a significant improvement in positive symptoms was observed. The patient has since had no delusions or hallucinations and is living independently at home. Contemporary research studies indicate that the lifetime rate of completed suicide in individuals with schizophrenia is between 4% and 13%. Several specific risk factors have been described in the schizophrenia population, such as early stage of the illness, lack of adherence to treatment, recurrent relapses, comorbid depression and the paranoid subtype. The antipsychotic treatments with the most scientific evidence are clozapine, risperidone, olanzapine and quetiapine. Within psychotherapy, cognitive behavioral therapy appears to be the most effective. Conclusions: -It is important to know the risk factors that are associated with an increased risk of suicide in patients diagnosed with schizophrenia. -An early intervention and specific treatment can improve prognosis of this population.

EPV1076
Werther effect in the juvenile population. About a series of cases Introduction: Suicide is currently one of the biggest public health problems, it is the third cause of death in the age group between 15 and 29 years (16.36% of young people who died in 2013). The 'Werther effect' refers to the mimetic behavior of the suicidal act, thus making reference to the controversial novel "The Sorrows of Young Werther" by Goethe, in 1774. The population most susceptible to this influence is the most vulnerable and ambivalent, such as they can be adolescents and young people, people with personality disorders and drug use. Durkheim considered that imitation was not due to the contagion effect of making suicides public, but to the social conditions of some places, which were what caused people to commit suicide.