Transfer learning with chest X-rays for ER patient classification

Abstract

One of the challenges with urgent evaluation of patients with acute respiratory distress syndrome (ARDS) in the emergency room (ER) is distinguishing between cardiac vs infectious etiologies for their pulmonary findings. We conducted a retrospective study with the collected data of 171 ER patients. ER patient classification for cardiac and infection causes was evaluated with clinical data and chest X-ray image data. We show that a deep-learning model trained with an external image data set can be used to extract image features and improve the classification accuracy of a data set that does not contain enough image data to train a deep-learning model. An analysis of clinical feature importance was performed to identify the most important clinical features for ER patient classification. The current model is publicly available with an interface at the web link: http://nbttranslationalresearch.org/ .

Figure 1

Examples of chest X-ray images where the ground truth label is (a) neither infection nor cardiac label, (b) infection label, (c) cardiac label.

Figure 2

Summary of fivefold cross validation results. Infection range (light)/fold (dark) performance is shown in blue. Cardiac range (light)/fold (dark) performance is shown in red. Black horizontal bars denote the mean across all folds.

Figure 3

SHAP TreeExplainer Feature importance plot for the top ten clinical features for the “cardiac” task (left) and the “infection” task (right). Color denotes feature magnitude, X-axis shows SHAP value, feature names are shown to the side of each row.

Figure 4

SHAP Feature importance plot for the top ten image features for the “cardiac” task (left) and the “infection” task (right). Color denotes feature magnitude, X-axis shows SHAP value, feature names are shown to the side of each row.