PFI-RP: Improving Inhaler Design and Efficacy with a Novel AI-assisted Digital Human Testing Platform

这一创新-研究伙伴关系(PFI-RP)项目的更广泛影响/商业潜力是提供了一个用于吸入器创新的虚拟人体测试平台，提高了药物输送效率，从而更好地治疗肺部疾病。该项目旨在完善和完善一个可以量化各种类型的吸入器在健康和患病肺部排放的气流和雾化药物的吸入，以及通过非侵入性、成本效益和高效的方式在人体内转移药物的“一体式虚拟人体测试平台”。虚拟测试平台预计将为吸入器创新和审批节省约66%的成本和90%的时间。因此，该项目可以改善向小气道的靶向药物输送，潜在地减少炎症并改善慢性肺部疾病患者的肺健康。这一虚拟测试平台的商业化还将加速制药公司在吸入器设计工程方面的创新，并促进肺部保健研究。拟议的项目将通过以下方式开发一个虚拟人体测试平台：(1)在人工智能(AI)和计算肺气溶胶动力学的辅助下，集成疾病特定的代表性全肺几何形状；(2)实现多种类型吸入器的液滴-蒸汽相互作用模拟；以及(3)为该平台设计易于导航的基于网络的用户界面(UI)，使得能够以广泛可访问的方式输入所需的药物配方和吸入器设计参数，并下载虚拟人体测试平台预测的生物等效性数据，以评估和优化吸入器设计，而不需要计算流体-颗粒动力学方面的专业知识。该项目还将通过开发一种人工智能辅助方法来构建具有代表性的慢性阻塞性肺疾病(COPD)全肺模型，从而促进吸入器创新方面的科学知识，这将帮助制药公司了解吸入器如何向疾病特定的肺环境提供准确的剂量。该项目还将作为一种非侵入性和成本效益高的替代工具，用于对患有不同肺部疾病的人群进行吸入器工程和设计的比较生物等效性研究。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The broader impact/commercial potential of this Partnerships for Innovation - Research Partnerships (PFI-RP) project is the availability of a virtual human testing platform for inhaler innovation, enhancing drug delivery efficiency for better lung disease treatment. The project is intended to refine and improve an “all-in-one virtual human testing platform” that can quantify airflow and the inhalation of aerosolized medication emitted from various types of inhalers in healthy and diseased lungs, as well as drug translocations in human body via a non-invasive, cost-effective, and efficient manner. The virtual testing platform is expected to save approximately 66% of the cost and 90% of the time for inhaler innovation and approval. Therefore, the project could improve the targeted drug delivery to small airways, potentially reducing inflammation and improving the pulmonary health of patients with chronic lung diseases. The commercialization of this virtual testing platform may also accelerate innovations in inhalation device design engineering for pharmaceutical companies and promote pulmonary healthcare research.The proposed project will develop a virtual human testing platform by: (1) integrating disease-specific representative whole-lung geometries with the assistance of artificial intelligence (AI) and computational lung aerosol dynamics; (2) enabling droplet-vapor interaction simulation for multiple types of inhalers; and (3) designing an easy-to-navigate web-based user interface (UI) for this platform, enabling a widely-accessible way to input desired drug formulation and inhaler design parameters, and downloading bioequivalence data predicted by the virtual human testing platform for evaluating and optimizing inhaler design without the need for expertise in computational fluid-particle dynamics. The project will also advance the scientific knowledge in inhaler innovation by developing an AI-assisted approach to construct representative disease-specific chronic obstructive pulmonary disease (COPD) whole-lung models, which will help pharmaceutical companies understand how inhalers can deliver accurate doses into the disease-specific lung environment. The project will also serve as a non-invasive and cost-effective alternative tool for the comparative bioequivalence study of inhaler engineering and design on population groups with different lung diseases.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.