PFI-TT: Prototyping a mobile app and paper test card system to detect dangerous opioids and stimulants in the field

这一创新-技术转化伙伴关系项目更广泛的影响/商业潜力是创建一个用于推定检测非法类阿片和兴奋剂的功能原型系统。这项新技术可以防止美国每年45，000例致命的阿片类药物过量，并减少虚假药物逮捕的机会，这对某些少数群体的影响不成比例。该技术使用一个移动的手机应用程序，再加上一张纸测试卡和机器视觉算法。 该卡片可以同时对样品进行12次化学测试，产生一个彩色条形码，显示样品的化学成分。移动的应用程序将使用机器视觉软件来读取测试结果。 该系统将比目前市售的推定药物检测更不可能将无害物质错误地识别为非法药物，并可能能够测量药物浓度和识别混合物中存在的多种药物。 后者是当前假定测试做得很差的能力，如果有的话。拟议的项目将开发一个非法药物识别系统原型，以供进一步商业开发。纸卡将配有剂量计量装置和溶剂储存器，以方便现场使用。millifluidic卡上的颜色测试数量将增加50%，针对阿片类药物和兴奋剂药物。 人类感知颜色的能力是主观的，并且可能受到色觉差异的影响，因此“通过眼睛”阅读测试卡结果的分析度量将与使用主成分分析、神经网络分类方法或生成对抗网络（GAN）方法的结果进行比较。无害的兴奋剂药物将用于现场测试，以评估不同的原型卡设计，并证明移动的应用程序+卡系统的潜力，用于对导致过量爆发的“产品”进行地理时间检测。该团队将使用商业模式画布框架和定期小组会议，以确保所有参与者接受创业培训，并确保研究和开发活动的结果是一个可商业化的系统，用于在现场检测非法药物。该奖项反映了NSF的法定使命，并已被认为是值得通过使用基金会的智力价值和更广泛的影响审查标准进行评估的支持。

The broader impact/commercial potential of this Partnerships for Innovation - Technology Translation (PFI-TT) project is to create a functional prototype system for presumptive detection of illicit opioids and stimulants. This new technology could prevent some of the 45,000 fatal opioid overdoses per year in the US and reduce the chances of false drug arrests, which disproportionately affect certain minority populations. The technology uses a mobile phone app coupled with a paper test card and machine vision algorithms. The paper card runs twelve chemical tests on the sample at once, producing a color bar code that reveals the chemical composition of the sample. The mobile app will use machine vision software to read the results of the test. The system will be less likely to erroneously identify a harmless substance as an illicit drug than current commercially available presumptive drug tests and may be able to measure drug concentrations and identify multiple drugs present in mixtures. The latter are capabilities that the current presumptive tests do poorly, if at all. The proposed project will develop a prototype illicit drug identification system for further commercial development. The paper card will be accessorized with a dose metering device and solvent reservoir to facilitate field use. The number of color tests on the millifluidic card will be increased by 50%, targeting opioids and stimulant drugs. Human ability to sense colors is subjective and may be affected by differences in color vision, so the analytical metrics of reading test card results “by eye” will be compared with the results of using principal component analysis, a neural network classification approach, or a generative adversarial network (GAN) approach. Harmless stimulant drugs will be used in field tests to evaluate different prototype card designs, and to demonstrate the potential of the mobile app + card system for geotemporal detection of a “product” responsible for an overdose outbreak. The team will use a business model canvas framework and regular group meetings to ensure that all participants receive training in entrepreneurship and that the research and development activities result in a commercializable system for detecting illicit drugs in field settings.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.