Identifying neural fingerprints of suicidality

在过去的20年里，自杀死亡人数一直在稳步上升，社会孤立和经济压力 不幸的是，与当前的大流行相关的可能为戏剧性的 自杀率上升。这种风险在退伍军人中更高，尤其是那些患有创伤性脑损伤的人。 和精神诊断，目前的公共卫生危机对心理健康具有令人震惊的影响。 目前的自杀预防实践在很大程度上是通过对自杀念头的评估和 行为(STB)和其他临床特征。自杀风险和预防的一个重要限制 完全依赖自我报告，而自我报告在预测未来自杀方面的有效性受到严重限制 企图和死亡，并没有确定谁没有透露想法或自残行为。至 测试一种替代当前预防模式的方法，我们建议补充神经成像- 基于自杀风险的生物标记物可以提高高危个体的识别能力。 设计与方法。我们的实验室在应用认知神经科学工具方面处于领先地位 精确精神病学。我们通过获取功能核磁共振来实现这一点，众所周知功能核磁共振是可持续重现的 在个人内部，但受制于个人之间的巨大变异性，使其对个人也是独特的 作为他们的神经精神和神经认知特征。在本提案中，我们将使用这些扫描来解析 通过大规模网络的大脑连接，包括情绪和抑制控制电路，这些电路 与性传播疾病有关。然后，通过应用机器学习技术，我们将分离大脑的模式 识别有自杀倾向的个人的活动。此外，我们将通过以下方式验证STB的这些神经标记物 从有自杀风险的退伍军人那里收集新的fMRI数据，同时进行自杀隐含关联 测试(S-IAT)，是一种客观的行为测量，已知可以预测未来的自杀企图。最后，我们会 确定STB的这些神经标记物是否也与日常和社会功能受损有关 STB的贡献者。这将是利用这些方法实现以下目标的首批研究之一 识别有自杀风险的个人。拟议的研究将利用现有的 脑损伤和应激障碍翻译研究中心的神经成像和临床数据也是如此 作为正在进行的数据收集，另外60名退伍军人将完成S-IAT的同步功能磁共振成像。 目标。目的1：开发一种基于神经成像的模型来检测当前有自杀倾向的个体 有自杀意念和/或自杀未遂史(S)。假设1.模型将区分有自杀倾向的个体 来自那些没有自杀倾向但有类似精神健康状况的人，基于功能性 与情绪调节和抑制性控制相关的大脑区域之间的连接。 目标2：确定目标1的横截面模型是否能预测哪些人会尝试 在接下来的1-2年内自杀。假设2.模型将确定至少50%的个人将 有类似心理健康状况但不会尝试自杀的人尝试自杀。 目的3：确定STB神经标志物的表达是否与功能减退有关 结果。假设3：STB的神经标记物将与降低的功能结局相关。 目的4：应用S放射免疫法测定STB的fMRI激活标志物。假设4.患有性病的退伍军人 在“我”和“死亡”之间有更高的关联性，同时大脑相关区域的激活程度也更高 有情绪调节和抑制性控制。 冲击力。成功识别自杀的神经特征将消除对自我的依赖 自杀念头的披露，并对准确识别和 防止自杀。未来的研究将集中在这些生物标记物在独立的， 预期样本，以及针对这些生物标志物的电路特异性脑刺激干预。

Death by suicide has been steadily increasing in the last 20 years, and the social isolation and financial stress associated with the current pandemic may unfortunately provide the perfect conditions for a dramatic increase in suicidality. This risk is elevated among veterans, particularly those with traumatic brain injury and psychiatric diagnoses, and the current public health crisis has alarming implications for mental health. Current suicide prevention practices are largely informed by the evaluation of suicidal thoughts and behaviors (STBs), and other clinical characteristics. One significant limitation in suicide risk and prevention is the exclusive reliance on self-report, which is severely limited in its effectiveness to predict future suicide attempts and deaths, and does not identify individuals who do not disclose thoughts or acts of self-harm. To test an alternative to current modes of prevention, we propose that complementary neuroimaging- based biomarkers of suicide risk can improve the identification of at-risk individuals. DESIGN AND METHODS. Our lab is a leader in the application of cognitive neuroscience tools toward precision psychiatry. We accomplish this by acquiring functional MRI, known to be consistently reproducible within an individual but subject to great variability across individuals, making it unique to the person, as well as their neuropsychiatric and neurocognitive profile. In this proposal, we will use these scans to parse out brain connections across large-scale networks including emotional and inhibitory control circuitry that are implicated in STBs. Then, by applying machine learning techniques, we will isolate the pattern of brain activity that identifies suicidal individuals. Further, we will validate these neural markers of STBs by collecting new fMRI data from veterans at risk for suicide while they perform the Suicide Implicit Association Test (S-IAT), an objective behavioral measure known to predict future suicide attempt. Finally, we will determine if these neural markers of STBs are also associated with impaired daily and social functioning, a contributor to STBs. This will be one of the first studies to leverage these methods towards the goal of identifying individuals at risk for suicide. The proposed study will accomplish these aims using both existing neuroimaging and clinical data from the Translational Research Center for TBI and Stress Disorders, as well as ongoing data collection in which 60 additional veterans will complete the S-IAT with concurrent fMRI. OBJECTIVES. Aim 1: Develop a neuroimaging-based model to detect individuals with current suicidal ideation and/or a history of suicide attempt(s). Hypothesis 1. Model will distinguish suicidal individuals from those who are not suicidal but who have comparable mental health conditions, based on functional connectivity between brain regions associated with emotional regulation and inhibitory control. Aim 2: Determine if the cross-sectional model from Aim 1 can predict which individuals will attempt suicide in the next 1-2 years. Hypothesis 2. Model will identify at least 50% of individuals who will attempt suicide from those with comparable mental health conditions who will not attempt suicide. Aim 3: Determine if the expression of the STB neural markers is associated with reduced functional outcomes. Hypothesis 3. Neural markers of STBs will be associated with reduced functional outcomes. Aim 4: Determine fMRI activation-based markers of STBs using the S-IAT. Hypothesis 4. Veterans with STBs will have higher associations between “me” and “death” alongside greater activation in brain regions associated with emotional regulation and inhibitory control. IMPACT. Successful identification of a neural signature of suicidality would remove reliance upon self- disclosure of suicidal thoughts and have dramatic clinical impact upon the precise identification and prevention of suicidality. Future studies will then focus on validation of these biomarkers in independent, prospective samples, as well as circuit-specific brain stimulation interventions targeting these biomarkers.