Stork: Advancing a novel portable detection method for cannabis intoxication

MELISSA CULHANE MARAVIC (2017-08-15 to 2018-07-31) Advancing a novel portable detection method for cannabis intoxication. Amount: $2084473



Intoxication from marijuana (MJ) impairs psychomotor performance and at least doubles the risk of motor vehicle accidents. The ongoing wave of legalization of MJ has brought increasing prevalence of driving while intoxicated with MJ. However, there is no quantitative biologic test that can accurately determine whether an individual is acutely impaired from MJ intoxication. Assays of the primary intoxicating substance in MJ, THC, in body fluids has a high false negative rate as THC is cleared from blood within 15 minutes, long before impairment is resolved. And assays of THC metabolites yield a high false positive rate because clearance of these metabolites can take weeks. Thus there is now no nor is there likely to ever be a test of blood, breath or body fluids that can accurately detect MJ intoxication. In response to this significant knowledge gap, this project aims to develop an accurate, portable method for detection of impairment due to MJ intoxication using functional near-infrared spectroscopy (fNIRS). fNIRS is a non-invasive, safe brain imaging technique that capitalizes on differences in the light absorption spectra of deoxygenated and oxygenated hemoglobin (Hb), that allows the measurement of relative changes in Hb concentration that reflect brain activity. fNIRS can be performed in natural environments at low cost, and thus can be used in real-world settings. In Phase I, we will develop an algorithm for individual-level detection of impairment from THC using fNIRS measurements. To do so, we will assess the effect of oral THC (or placebo) on fNIRS measurements, self-reported intoxication, and impairment as defined by the gold standard field sobriety test conducted by a Drug Recognition Expert (DRE) in 40 healthy MJ users. fNIRS assessments will examine (1) the effect of THC exposure on resting state and task-based activation in the prefrontal cortex, (2) the extent to which impairment in psychomotor functioning with THC administration correlates with THC-induced change in hemodynamic responses detected with fNIRS, and (3) the sensitivity and specificity and area under the ROC curve of fNIRS measurements and field sobriety test determinants of impairment. Milestone: Should machine learning applications to the data generate an algorithm that predicts impairment with >80% accuracy compared with a gold standard field sobriety test, we will proceed to Phase II. In Phase II, we will conduct fNIRS testing in 150 individuals under THC/placebo as in Phase I and in 50 individuals in a THC plus alcohol/placebo condition in order to further refine the algorithm for MJ impairment detection such that fNIRS detection concurs with field sobriety testing with >90% specificity. It is anticipated that this level of specificity could be used in legal definitions of impairment. This will warrant commercialization, which will be followed by prototype development and field testing. An accurate, quantitative, biological test that is user-friendly and enables law enforcement to detect impairment from MJ has the potential to dramatically change practice of law enforcement across the country and the world and thus has enormous commercial potential, as outlined in the Commercialization Plan and in accompanying letters of support.

大麻中毒(MJ)会损害精神运动表现,至少使机动车事故的风险增加一倍。正在进行的MJ合法化浪潮使得驾驶时的驾驶越来越频繁,同时让MJ陶醉。然而,没有定量生物学测试可以准确地确定个体是否因MJ中毒而急性受损。 MJ,THC中的主要中毒物质在体液中的测定具有高的假阴性率,因为THC在15分钟内从血液中清除,早在损害解决之前。并且THC代谢物的测定产生高假阳性率,因为这些代谢物的清除可能需要数周。因此,现在没有也没有可能测试血液,呼吸或体液可以准确地检测MJ中毒。为了应对这一重要的知识差距,该项目旨在开发一种准确,便携的方法,使用功能性近红外光谱(fNIRS)检测MJ中毒引起的损伤。 fNIRS是一种非侵入性,安全的脑成像技术,可充分利用脱氧和氧合血红蛋白(Hb)的光吸收光谱差异,从而可以测量反映大脑活动的Hb浓度的相对变化。 fNIRS可以在自然环境中以低成本执行,因此可以在现实环境中使用。在第一阶段,我们将开发一种使用fNIRS测量从THC进行个体水平检测的算法。为此,我们将评估口服THC(或安慰剂)对fNIRS测量,自我报告中毒和损伤的影响,如由药物识别专家(DRE)在40名健康MJ用户中进行的金标准野外清醒测试所定义的。 fNIRS评估将检查(1)THC暴露对前额皮质静息状态和基于任务的激活的影响,(2)THC给药时精神运动功能受损的程度与THC诱导的血流动力学反应变化相关。 fNIRS,和(3)fNIRS测量和现场清醒测试的损伤决定因素的ROC曲线下的敏感性和特异性以及面积。里程碑:与金标准现场清醒测试相比,机器学习应用程序应该生成一种能够以80%以上的精度预测损伤的算法,我们将进入第二阶段。在第二阶段,我们将在THC /安慰剂的150名个体中进行fNIRS测试(如第一阶段)和50名个体的THC加酒精/安慰剂条件,以进一步改进MJ损伤检测算法,使fNIRS检测与野外一致清醒测试,特异性> 90%。预计这种特异性水平可用于减损的法律定义。这将保证商业化,随后将进行原型开发和现场测试。准确,定量,生物学测试,用户友好,并使执法部门能够发现MJ的损害,有可能大大改变全国和全世界的执法实践,因此具有巨大的商业潜力,如商业化计划所述并在随附的支持信中。

■Do you need the full text of this grant application? We can help you to apply for it. The fee is $150 (USD). Please write to us with subject line Full text request for grant R42DA043977 (NIDA). Please understand: (1) the application process involves NIH and the original grant authors and it will take ~1 month; (2) the grant author may exclude some sensitive information from the full text.