Quantitative estimation of pulmonary artery wedge pressure from chest radiographs by a regression convolutional neural network

Abstract

Recent studies reported that a convolutional neural network (CNN; a deep learning model) can detect elevated pulmonary artery wedge pressure (PAWP) from chest radiographs, the diagnostic images most commonly used for assessing pulmonary congestion in heart failure. However, no method has been published for quantitatively estimating PAWP from such radiographs. We hypothesized that a regression CNN, an alternative type of deep learning, could be a useful tool for quantitatively estimating PAWP in cardiovascular diseases. We retrospectively enrolled 936 patients with cardiovascular diseases who had undergone right heart catheterization (RHC) and chest radiography and estimated PAWP by constructing a regression CNN based on the VGG16 model. We randomly categorized 80% of the data as training data (training group, n = 748) and 20% as test data (test group, n = 188). Moreover, we tuned the learning rate—one of the model parameters—by 5-hold cross-validation of the training group. Correlations between PAWP measured by RHC [ground truth (GT) PAWP] and PAWP derived from the regression CNN (estimated PAWP) were tested. To visualize how the regression CNN assessed the images, we created a regression activation map (RAM), a visualization technique for regression CNN. Estimated PAWP correlated significantly with GT PAWP in both the training (r = 0.76, P < 0.001) and test group (r = 0.62, P < 0.001). Bland–Altman plots found a mean (SEM) difference between GT and estimated PAWP of − 0.23 (0.16) mm Hg in the training and − 0.05 (0.41) mm Hg in the test group. The RAM showed that our regression CNN model estimated high PAWP by focusing on the cardiomegaly and pulmonary congestion. In the test group, the area under the curve (AUC) for detecting elevated PAWP (≥ 18 mm Hg) produced by the regression CNN model was similar to the AUC of an experienced cardiologist (0.86 vs 0.83, respectively; P = 0.24). This proof-of-concept study shows that regression CNN can quantitatively estimate PAWP from standard chest radiographs in cardiovascular diseases.

Reliability Metrics of Explainable CNN based on Wasserstein Distance for Cardiac Evaluation

“…In this study, an all-analysis procedure, we used this R-CNN and the same dataset from [6]. We confirmed almost all RAMs generated by the R-CNN contained the cardiac region (see Appendix section) using the k-means clustering approach.…”

“…The size of the RAM was 8 × 8, and was resized to 256 × 256 (input image size). The details of other training conditions are explained in [6].…”

“…• We used the dataset described in a previous study [6]. The data were collected via a single center study.…”

“…Therefore, it is important to develop an R-CNN, which outputs RAMs containing only the cardiac region. As for future works, we plan to develop such an R-CNN by improving the previous model [6].…”

“…Therefore, an objective method, convolutional neural network (CNN), has been reported for estimating PAWP from chest radiograph. For example, Hirata et al [5] developed a classification CNN to detect PAWP over 18 mmHg, and Saito et al [6] developed a regression CNN to quantitatively estimate PAWP from chest radiographs. Because these CNNs output a class activation map (CAM) [7] or regression activation map (RAM) [8], physicians can detect PAWP and estimate the activated regions, non-invasively.…”