Huaiwu He, MD | Chinese Academy of Medical Sciences; Siyi Yuan, MD; Yi Chi, MD; Zhanqi Zhao, PhD; Maokun Li, PhD
Competition Sponsor: Chinese Academy of Medical Sciences
Awardee Year: 2025
Acute respiratory failure (ARF) caused by conditions such as pneumonia or pulmonary embolism is a major cause of mortality in critically ill patients, especially the elderly, and poses a significant threat to healthy longevity. Monitoring pulmonary ventilation and perfusion is central to ARF management; however, conventional tools such as chest X-ray, CT, CTPA, and nuclear imaging carry transport risks, radiation exposure, and procedural delays, making them unsuitable for real-time bedside assessment. Thus, a noninvasive, bedside, and dynamic monitoring technique is urgently needed to support precision treatment of ARF. This project aims to develop a high-precision bedside three-dimensional electrical impedance tomography (3D-EIT) technology for simultaneous imaging of lung ventilation and perfusion. By integrating individualized 4D-CT–based dynamic calibration with AI-driven algorithms, the system will enable real-time and accurate whole-lung monitoring, while clinical studies will establish evidence-based strategies for its application in respiratory failure management. The technological pathway includes addressing the nonlinear nature of 3D-EIT using physics-informed neural networks (PINN) for forward physical modeling and embedding physical constraints into network optimization; incorporating patient-specific 3D-CT lung contours and 4D-CT ventilation dynamics as structural priors to mitigate the ill-posedness of 3D-EIT reconstruction; leveraging large-scale GPU parallel computing to enable real-time 3D image reconstruction; and conducting clinical research to develop practical clinical implementation strategies for 3D-EIT.