No abstract

COVID-19 is spreading around the world like wildfire. Chest X-rays are used as one of the primary tools for diagnosing COVID-19. However, about two-thirds of the world population do not have access to sufficient radiological services. In this work, we propose a deep learning-driven automated system, COVIDXception-Net, for diagnosing COVID-19 from chest X-rays. A primary challenge in any data-driven COVID-19 detection is the scarcity of COVID-19 data, which heavily deteriorates a deep learning model’s performance. To address this issue, we incorporate a weighted-loss function that ensures the COVID-19 cases are given more importance during the training process. We also propose using Bayesian Optimization to find the best architecture for detecting COVID-19. Extensive experimentation on four publicly available COVID-19 datasets shows that our proposed model achieves an accuracy of 0.94, precision 0.95, recall 0.94, specificity 0.997, F1-score 0.94, and Matthews correlation coefficient 0.992 outperforming three widely used architectures—VGG16, MobileNetV2, and InceptionV3. It also surpasses the performance of several state-of-the-art COVID-19 detection methods. We also performed two ablation studies that show our model’s accuracy degrades from 0.994 to 0.950 when a random search is used and to 0.983 when a regular loss function is employed instead of the Bayesian and weighted loss, respectively.

Neural Architecture Search (NAS) is the process of automating the design of neural network architectures for a given task. Although NAS automates the process of finding suitable neural network architectures for a specific task, the existing NAS algorithms are immensely time-consuming. The main bottleneck in NAS algorithms is the training time for each architecture. This study proposes an Improved Grey Wolf Optimization based on Synaptic Saliency (IGWO-SS), which is much faster than the existing NAS algorithms and provides better final performance. The IGWO-SS algorithm skips training the less promising architectures by creating a relative rank between the architectures based on synaptic saliency. The architectures that are lower in rank are considered less promising than those that are higher in rank. Since the calculation of synaptic saliency is a very fast process, a significant amount of time is saved by skipping training of less promising architectures. This study involves extensive experiments assessing synaptic saliency's effectiveness in improving NAS. The experimental results indicate that the synaptic saliency of an untrained neural network positively correlates with its final accuracy. Hence, it can be used to identify untrained promising neural networks. The experimental results suggest that the IGWO-SS algorithm is almost 10x faster and achieves better final performance than five other bio-inspired algorithms. The IGWO-SS algorithm achieves higher mean accuracy than state-of-the-art NAS algorithms, including -REA, RS, RL, BOHB, DARTSV1, DARTSV2, GDAS, SETN, and ENAS. We hope our work will make NAS more accessible and useful to researchers by reducing the time and resources required to perform NAS.INDEX TERMS neural architecture search; nas; grey wolf optimization; gwo; automl; deep learning

No abstract

No abstract