Exploration Of Interpretability Techniques For Deep COVID-19 Classification Using Chest X-Ray Images

Abstract

The outbreak of COVID-19 has shocked the entire world with its fairly rapid spread and has challenged different sectors. One of the most effective ways to limit its spread is the early and accurate diagnosis of infected patients. Medical imaging such as X-ray and Computed Tomography (CT) combined with the potential of Artificial Intelligence (AI) plays an essential role in supporting the medical staff in the diagnosis process. Thereby, five different deep learning models (ResNet18, ResNet34, InceptionV3, InceptionResNetV2, and DenseNet161) and their Ensemble have been used in this paper, to classify COVID-19, pneumoniae and healthy subjects using Chest X-Ray images. Multi-label classification was performed to predict multiple pathologies for each patient, if present. Foremost, the interpretability of each of the networks was thoroughly studied using techniques like occlusion, saliency, input X gradient, guided backpropagation, integrated gradients, and DeepLIFT. The mean Micro-F1 score of the models for COVID-19 classifications ranges from 0.66 to 0.875, and is 0.89 for the Ensemble of the network models. The qualitative results depicted the ResNets to be the most interpretable models.

A comparative analysis of eleven neural networks architectures for small datasets of lung images of COVID-19 patients toward improved clinical decisions

“…The images were then filled with a background composed of 10 translational and rotational lungs [24] NONE [26] The conventional data augmentation method included ± 15 • rotation, ± 15% x-axis shift, ± 15% y-axis shift, horizontal flipping, and 85%-115% scaling and shear transformation. The parameters of mixup was set to 0.1 [9] The original image is divided into 16 * 16 and 32 * 32 blocks to build two data sets [27] All the of images were initially preprocessed to have the same size. To make the image size uniform throughout the dataset, each of the images was interpolated using bicubic interpolation.…”

“…Performance Criteria [8] AUC\Recall\Precision\F1-score\Accuracy [24] Accuracy\Sensitivity\FPR\F1-score [26] Accuracy\Sensitivity [9] TP\TN\FP\FN\Accuracy \Sensitivity\Specificity\Precision \F1-score \Matthews Correlation Coefficient (MCC) [27] \F1-score\Recall\Precision\Specificity [28] AUC\Recall\Precision\F1-score\Accuracy [29] AUC\Recall\Precision\F1-score\Accuracy [11] AUC\Sensitivity\Specificity [12] Recall\Precision\F1-score [13] AUC\Sensitivity\Specificity [16] AUC\Recall\Precision\Accuracy [21] AUC\specificity\Precision\F1-score\Accuracy value, the optimal historical value is updated. At the same time, the optimal weight file for this generation of training is saved.…”

A survey of machine learning techniques for detecting and diagnosing COVID‐19 from imaging

“…Grad-CAM [31,34,37,38,44,45,53,55,56,63,68,73,76,77,79,84,97] Grad-CAM++ [31,34,40] CAM [30,37,54,55,60,87,88,94] LIME [14,41,68] LRP [31] smallest. Next, for feature extraction (in Tables 4 and 5), most of the previous research has focused on the use of deep features, and the most widely used CNN architecture for feature extraction is ResNet.…”

“…[4,[15][16][17]19,20,22,[24][25][26][27][28]34,35,[37][38][39]41,43,[44][45][46]48,49,52,55,58,61,63,69,74,78,80,81,83,85,86,89,90,92,93,95,[99][100][101]105,107,111,[113][114][115] Multi-class[14,18,21,…”

A review of explainable and interpretable AI with applications in COVID‐19 imaging

“…Example heatmaps obtained from a variety of techniques (top) 70 and from Grad‐CAM (bottom) 67 . Each technique may provide different evaluations of influential regions, both in terms of relative importance and key locations.…”