2020/10/19 by Nils Strodthoff, Strodthoff, Nils, Claas Strodthoff +7
Engineering · Medicine · #Electrical and Bioimpedance Tomography #FOS: Computer and information sciences #FOS: Electrical engineering #Hemodynamic Monitoring and Therapy #Image and Video Processing (eess.IV) #Intravenous Infusion Technology and Safety #Machine Learning (cs.LG) #Signal Processing (eess.SP) #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2010.09622
openalex publication_date 2020/10/19 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
Electrical impedance tomography (EIT) is a noninvasive imaging modality that\nallows a continuous assessment of changes in regional bioimpedance of different\norgans. One of its most common biomedical applications is monitoring regional\nventilation distribution in critically ill patients treated in intensive care\nunits. In this work, we put forward a proof-of-principle study that\ndemonstrates how one can reconstruct synchronously measured respiratory or\ncirculatory parameters from the EIT image sequence using a deep learning model\ntrained in an end-to-end fashion. We demonstrate that one can accurately infer\nabsolute volume, absolute flow, normalized airway pressure and within certain\nlimitations even the normalized arterial blood pressure from the EIT signal\nalone, in a way that generalizes to unseen patients without prior calibration.\nAs an outlook with direct clinical relevance, we furthermore demonstrate the\nfeasibility of reconstructing the absolute transpulmonary pressure from a\ncombination of EIT and absolute airway pressure, as a way to potentially\nreplace the invasive measurement of esophageal pressure. With these results, we\nhope to stimulate further studies building on the framework put forward in this\nwork.\n