MedCheX: Automatic AI-based Pneumonia Detection from Chest X-ray Images and It’s Application to Coronavirus Pandemic (COVID-19)
Automatic detection of pneumonia on X-rays continues to be an important issue. Our system will process the images to assist doctors to determine whether the patients are infected. We first created a ResNeXt based deep learning model to automatically detect the presence of pneumonia. For enhancing the ability of feature extraction in the model, we re-designed the model structure by using the squeeze-and-excitation, which can extract more fine-grained features in the process. Meanwhile, we utilized the Feature Pyramid Network to extend the receptive field on the convolutional kernel, which improved the performance on the pneumonia detection with various location on CXR. In the training phase, we applied the data augmentation, included flipping, rotation, adding noises, and image distortion, to make the model more robust and avoid the model overfitting.
As we continue to face the rapid increase in confirmed Coronavirus cases around the world, we created an AI-based pneumonia detection platform for COVID-19. The system is able to automatically detect high-risk patients with pneumonia that will then send alert information to doctors. With that information, the doctors are then able to make follow-up decisions and provide a treatment plan after the diagnosis. In specific, doctors from Department of Medical Imaging provided us thousands of positive and negative chest x-rays for pneumonia as training set. Our system has already been tested with and adopted by doctors at the NCKU Hospital. The system achieved 95.5% sensitivity and 99.0% specificity to detect the pneumonia symptom, based on 1,363 test images.