Deep Learning-Based Detection of the Acute Respiratory Distress Syndrome: What Are the Models Learning?
This work addresses the challenge of prompt ARDS diagnosis to reduce high mortality rates in hospitals, but it is incremental as it builds on existing deep learning methods for medical data.
The paper tackled the problem of detecting acute respiratory distress syndrome (ARDS) using deep learning on ventilator waveform data, resulting in a convolutional neural network model that outperformed prior random forest models with an AUC of 0.95±0.019 vs. 0.88±0.064, accuracy of 0.84±0.026 vs. 0.80±0.078, and specificity of 0.81±0.06 vs. 0.71±0.089.
The acute respiratory distress syndrome (ARDS) is a severe form of hypoxemic respiratory failure with in-hospital mortality of 35-46%. High mortality is thought to be related in part to challenges in making a prompt diagnosis, which may in turn delay implementation of evidence-based therapies. A deep neural network (DNN) algorithm utilizing unbiased ventilator waveform data (VWD) may help to improve screening for ARDS. We first show that a convolutional neural network-based ARDS detection model can outperform prior work with random forest models in AUC (0.95+/-0.019 vs. 0.88+/-0.064), accuracy (0.84+/-0.026 vs 0.80+/-0.078), and specificity (0.81+/-0.06 vs 0.71+/-0.089). Frequency ablation studies imply that our model can learn features from low frequency domains typically used for expert feature engineering, and high-frequency information that may be difficult to manually featurize. Further experiments suggest that subtle, high-frequency components of physiologic signals may explain the superior performance of DL models over traditional ML when using physiologic waveform data. Our observations may enable improved interpretability of DL-based physiologic models and may improve the understanding of how high-frequency information in physiologic data impacts the performance our DL model.