Currently, most mask extraction techniques are based on convolutional neural networks (CNNs). However, there are still numerous problems that mask extraction techniques need to solve. Thus, the most advanced methods to deploy artificial intelligence (AI) techniques are necessary. The use of cooperative agents in mask extraction increases the efficiency of automatic image segmentation. Hence, we introduce a new mask extraction method that is based on multi-agent deep reinforcement learning (DRL) to minimize the long-term manual mask extraction and to enhance medical image segmentation frameworks. A DRL-based method is introduced to deal with mask extraction issues. This new method utilizes a modified version of the Deep Q-Network to enable the mask detector to select masks from the image studied. Based on COVID-19 computed tomography (CT) images, we used DRL mask extraction-based techniques to extract visual features of COVID-19 infected areas and provide an accurate clinical diagnosis while optimizing the pathogenic diagnostic test and saving time. We collected CT images of different cases (normal chest CT, pneumonia, typical viral cases, and cases of COVID-19). Experimental validation achieved a precision of 97.12% with a Dice of 80.81%, a sensitivity of 79.97%, a specificity of 99.48%, a precision of 85.21%, an F1 score of 83.01%, a structural metric of 84.38%, and a mean absolute error of 0.86%. Additionally, the results of the visual segmentation clearly reflected the ground truth. The results reveal the proof of principle for using DRL to extract CT masks for an effective diagnosis of COVID-19.

Background: Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) is a global threat impacting the lives of millions of people worldwide. Automated detection of lung infections from Computed Tomography scans represents an excellent alternative; however, segmenting infected regions from CT slices encounters many challenges. Objective: Developing a diagnosis system based on deep learning techniques to detect and quantify COVID-19 infection and pneumonia screening using CT imaging. Method: Contrast Limited Adaptive Histogram Equalization pre-processing method was used to remove the noise and intensity in homogeneity. Black slices were also removed to crop only the region of interest containing the lungs. A U-net architecture, based on CNN encoder and CNN decoder approaches, is then introduced for a fast and precise image segmentation to obtain the lung and infection segmentation models. For better estimation of skill on unseen data, a fourfold cross-validation as a resampling procedure has been used. A three-layered CNN architecture, with additional fully connected layers followed by a Softmax layer, was used for classification. Lung and infection volumes have been reconstructed to allow volume ratio computing and obtain infection rate. Results: Starting with the 20 CT scan cases, data has been divided into 70% for the training dataset and 30% for the validation dataset. Experimental results demonstrated that the proposed system achieves a dice score of 0.98 and 0.91 for the lung and infection segmentation tasks, respectively, and an accuracy of 0.98 for the classification task. Conclusions: The proposed workflow aimed at obtaining good performances for the different system’s components, and at the same time, dealing with reduced datasets used for training.