Abstract
Knowledge of gas volume, tissue mass and recruitability measured by the quantitative CT scan analysis (CT-qa) is important when setting the mechanical ventilation in acute respiratory distress syndrome (ARDS). Yet, the manual segmentation of the lung requires a considerable workload. Our goal was to provide an automatic, clinically applicable and reliable lung segmentation procedure. Therefore, a convolutional neural network (CNN) was used to train an artificial intelligence (AI) algorithm on 15 healthy subjects (1,302 slices), 100 ARDS patients (12,279 slices), and 20 COVID-19 (1,817 slices). Eighty percent of this populations was used for training, 20% for testing. The AI and manual segmentation at slice level were compared by intersection over union (IoU). The CT-qa variables were compared by regression and Bland Altman analysis. The AI-segmentation of a single patient required 5–10 s vs. 1–2 h of the manual. At slice level, the algorithm showed on the test set an IOU across all CT slices of 91.3 ± 10.0, 85.2 ± 13.9, and 84.7 ± 14.0%, and across all lung volumes of 96.3 ± 0.6, 88.9 ± 3.1, and 86.3 ± 6.5% for normal lungs, ARDS and COVID-19, respectively, with a U-shape in the performance: better in the lung middle region, worse at the apex and base. At patient level, on the test set, the total lung volume measured by AI and manual segmentation had a R2 of 0.99 and a bias −9.8 ml [CI: +56.0/−75.7 ml]. The recruitability measured with manual and AI-segmentation, as change in non-aerated tissue fraction had a bias of +0.3% [CI: +6.2/−5.5%] and −0.5% [CI: +2.3/−3.3%] expressed as change in well-aerated tissue fraction. The AI-powered lung segmentation provided fast and clinically reliable results. It is able to segment the lungs of seriously ill ARDS patients fully automatically.
Highlights
The quantitative analysis of lung tomography [quantitative CT scan analysis (CT-qa)] images has been used extensively for more than 20 years and has significantly improved our knowledge of the pathophysiology of the acute respiratory distress syndrome (ARDS; ARDS Definition Task Force et al, 2012)
We found that automatic lung segmentation performed by a properly trained neural network provided lung contours in close agreement with the ones obtained by manual segmentation
The trained model based on the U-Net can automatically segment the lungs in the CT with the limitations mentioned
Summary
The quantitative analysis of lung tomography [quantitative CT scan analysis (CT-qa)] images has been used extensively for more than 20 years and has significantly improved our knowledge of the pathophysiology of the acute respiratory distress syndrome (ARDS; ARDS Definition Task Force et al, 2012). With CT-qa we have clarified how densities are Automatic Lung Segmentation in CT distributed in ARDS, advancing the concept of the “baby lung” (Gattinoni et al, 1987; Bone, 1993; Gattinoni and Pesenti, 2005), showing how densities redistribute in prone position (Gattinoni et al, 1991; Pelosi et al, 1998; Cornejo et al, 2013), and explaining the mechanisms by which positive end-expiratory pressure (PEEP) acts (Pelosi et al, 1994). Determining the change in the non-aerated tissue fraction at two end-expiratory pressure levels, i.e., 5 and 45 cmH2O, is considered the gold standard for assessing recruitment in ARDS (Gattinoni et al, 2006). The actual segmentation procedure in several hospitals required constisten manual intervention. The time requirement and need of expert personal has serious hindered a broder adoption of CT-qa in clinical practice
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
