Towards an effective and efficient deep learning model for COVID-19 patterns detection in X-ray images

Abstract

Purpose Confronting the pandemic of COVID-19 is nowadays one of the most prominent challenges of the human species. A key factor in slowing down the virus propagation is the rapid diagnosis and isolation of infected patients. The standard method for COVID-19 identification, the Reverse transcription polymerase chain reaction method, is time-consuming and in short supply due to the pandemic. Thus, researchers have been looking for alternative screening methods, and deep learning applied to chest X-rays of patients has been showing promising results. Despite their success, the computational cost of these methods remains high, which imposes difficulties to their accessibility and availability. Thus, the main goal of this work is to propose an accurate yet efficient method in terms of memory and processing time for the problem of COVID-19 screening in chest X-rays. Methods To achieve the defined objective, we propose a new family of models based on the EfficientNet family of deep artificial neural networks which are known for their high accuracy and low footprints. We also exploit the underlying taxonomy of the problem with a hierarchical classifier. A dataset of 13,569 X-ray images divided into healthy, non-COVID-19 pneumonia, and COVID-19 patients is used to train the proposed approaches and other 5 competing architectures. We also propose a cross-dataset evaluation with a second dataset to evaluate the method generalization power. Results The results show that the proposed approach was able to produce a high-quality model, with an overall accuracy of 93.9%, COVID-19 sensitivity of 96.8%, and positive prediction of 100% while having from 5 to 30 times fewer parameters than the other tested architectures. Larger and more heterogeneous databases are still needed for validation before claiming that deep learning can assist physicians in the task of detecting COVID-19 in X-ray images, since the cross-dataset evaluation shows that even state-of-the-art models suffer from a lack of generalization power. Conclusions We believe the reported figures represent state-of-the-art results, both in terms of efficiency and effectiveness, for the COVIDx database, a database of 13,800 X-ray images, 183 of which are from patients affected by COVID-19. The current proposal is a promising candidate for embedding in medical equipment or even physicians' mobile phones.

COVID-Transformer: Interpretable COVID-19 Detection Using Vision Transformer for Healthcare

“…Accuracy Precision Recall F1 Score AUC Inception-V3 [31] 0.90 0.89 0.91 0.89 0.92 EfficientNet-B0 [30] 0.89 0.88 0.89 0.88 0.92 MobileNet-V2 [31] 0.90 0.90 0.89 0.90 0.92 ResNet-V2 [29,31] 0.88 0.87 0.86 0.86 0.93 VGG-16 [23,27,29,31] 0.87 0.87 0.85 0.86 0.90 Xception [24,31] 0.90 0.92 0.87 0.90 0.93 DenseNet-121 [25,29,31] 0.88 0.90 0.85 0.87 0.92 COVID-Transformer (Ours) 0.92 0.93 0.89 0.91 0.98…”

“…In [29], Xueyu et al fine-tuned multiple pre-trained models over a 2500 CT scan images data set and achieved 82.5% accuracy. Luz et al [30] proposed models based on the EfficientNet family with a hierarchical classifier and achieved an overall accuracy of 93.9%. Pham et al [31] presented a comprehensive study on transfer learning for COVID-19 detection from CT images by training and comparing 16 pre-trained models.…”

A deep transfer learning based model for automatic detection of COVID-19 from chest X-ray

“…GAN helped generate 30 times more images from the original dataset and helped overcome the overfitting problem by making the proposed model more robust. Luz et al [17] extended the EfficientNet model using chest x-ray images to perform COVID-19 detection. They reduced the number of parameters by a factor of 30 compared to the baseline research model and used 5 and 28 times less parameters than ResNet50 and VGG16, respectively.…”

CoWarriorNet: A Novel Deep-Learning Framework for CoVID-19 Detection from Chest X-Ray Images

“…Although the model achieved a higher accuracy, the segmentation and detection is a sequential process and not applied in the real-time scenario. Luz et al [28] used a pre-trained Efficient Net model evaluated on the CoVIDx dataset which also suffered the aforementioned problems of the pre-trained models along with data augmentation. Karakanis et al [29] obtained Generative Adversarial Network with ResNet8 as the discriminator which utilized transfer learning for realtime weight transfer, thereby deducing exponential space complexity.…”