I-Corps: Accessible Chemical Authentication of Counterfeit and Falsified Medicines

I-Corps项目的更广泛影响/商业潜力在于提供一种更快速和更具成本效益的替代方法，以化学方式鉴定实地的未知物质。通过将便携式化学测试卡和配套的移动的应用程序相结合，执法部门或制药部门的潜在客户可以随时在工作中实时测试未知物质。此外，该技术还可以更容易地从每个化学认证中挖掘大量数据。这有助于更全面地了解假冒产品在受影响地区的传播情况，从而保护整个社区免受潜在有害物质的侵害，同时使监管机构和政府机构能够采取有效措施打击假冒产品。这个I-Corps项目包括一个获得专利的化学测试卡和一个配套的移动的应用程序，这些应用程序已经经过了广泛的初步试验。纸分析设备（PAD）是一种廉价的化学测试卡，它可以使用比色分析进行快速的药物现场筛选。该设备目前可检测多达60种活性药物成分（API），这些活性药物成分通常存在于各种药物中，如抗疟药、抗生素和镇痛药，沿着辅料通常存在于假冒产品中，如粘合剂、填充剂和片剂涂层。Veripad移动的应用程序可使用机器学习算法对每个PAD进行即时分析。该应用程序使用深度卷积神经网络模型给出了反应的PAD卡与API匹配的概率。此外，该应用程序能够从使用各种图像特征提取和聚类算法（如ICA，PCA和K-means）生成的化学颜色变化的“手工制作”特征中执行化学成分分析。通过持续的研究，PAD和移动的应用程序有可能帮助各种药品的质量测试，最终打击假冒产品的传播。该奖项反映了NSF的法定使命，并被认为值得通过使用基金会的知识价值和更广泛的影响审查标准进行评估来支持。

The broader impact/commercial potential of this I-Corps project is in providing a more rapid and cost-effective alternative method to chemically authenticate unknown substances in the field. By combining a portable chemical test card and a companion mobile app, potential customers within law enforcement or pharmaceuticals can readily test unknown substances on the job in real-time. In addition, the technology makes it easier to mine massive amounts of data from each chemical authentication. This allows for a more comprehensive understanding of the spread of counterfeit products in affected regions, thereby protecting whole communities from potentially harmful substances while enabling regulatory and government agencies to take effective measures against counterfeiters. This I-Corps project includes a patented chemical test card and a companion mobile app that have gone through extensive preliminary trials. The Paper Analytical Device (PAD) is an inexpensive chemical test card which enables fast in-field screening of pharmaceuticals using colorimetric analysis. The device currently detects up to 60 active pharmaceutical ingredients (APIs) commonly found in a variety of medicines such as antimalarials, antibiotics, and analgesics, along with excipients that are commonly found in counterfeit products such as binders, fillers, and tablet coatings. The Veripad mobile application enables instantaneous analysis of each PAD using a machine learning algorithm. The app gives a probability of a reacted PAD card matching the API using deep convolutional neural network models. Moreover, the app is able to perform chemical composition analysis from the 'hand crafted' features of chemical color changes generated using various image feature extraction and clustering algorithms, such as ICA, PCA, and K-means. With ongoing research, the PAD and mobile app has the potential to aid in the quality testing of a variety of pharmaceuticals to ultimately combat the spread of counterfeits.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.