Currently used lung disease screening tools are expensive in terms of money and time. Therefore, chest radiograph images (CRIs) are employed for prompt and accurate COVID-19 identification. Recently, many researchers have applied deep learning (DL) based models to detect COVID-19 automatically. However, their model could have been more computationally expensive and less robust, i.e., its performance degrades when evaluated on other datasets. This study proposes a trustworthy, robust, and lightweight network (ChestCovidNet) that can detect COVID-19 by examining various CRI datasets. The ChestCovidNet model has only 11 learned layers, 8 convolutional (Conv) layers, and 3 fully connected (FC) layers. The framework employs both the Conv and group Conv layers, Leaky Relu activation function, shufflenet unit, Conv kernels of 3 × 3 and 1 × 1 to extract features at different scales, and two normalization procedures that are cross-channel normalization and batch normalization, which make it a novel COVID-19 identification framework. We used data augmentation techniques to increase the number of images and validate the generalizability of the suggested model. We used 9013 CRIs for training, whereas 3863 CRIs for testing the proposed ChestCovidNet approach. Furthermore, we compared the classification results of the proposed framework with hybrid methods in which we employed DL frameworks for feature extraction and support vector machines (SVM) for classification. The study’s findings demonstrated that the embedded low-power ChestCovidNet model worked well and achieved a classification accuracy of 98.12% and recall, F1-score, and precision of 95.75% for the four class classifications of CRIs (COVID-19, normal, pneumonia, and lung opacity (LO)).

Image encryption has attracted much interest as a robust security solution for preventing unauthorized access to critical image data. Medical picture encryption is a crucial step in many cloud-based and healthcare applications. In this study, a strong cryptosystem based on a 2D chaotic map and Jigsaw transformation is presented for the encryption of medical photos in private Internet of Medical Things (IoMT) and cloud storage. A disorganized three-dimensional map is the foundation of the proposed cipher. The dispersion of pixel values and the permutation of their places in this map are accomplished using a nonlinear encoding process. The suggested cryptosystem enhances the security of the delivered medical images by performing many operations. To validate the efficiency of the recommended cryptosystem, various medical image kinds are used, each with its unique characteristics. Several measures are used to evaluate the proposed cryptosystem, which all support its robust security. The simulation results confirm the supplied cryptosystem’s secrecy. Furthermore, it provides strong robustness and suggested protection standards for cloud service applications, healthcare, and IoMT. It is seen that the proposed 3D chaotic cryptosystem obtains an average entropy of 7.9998, which is near its most excellent value of 8, and a typical NPCR value of 99.62%, which is also near its extreme value of 99.60%. Moreover, the recommended cryptosystem outperforms conventional security systems across the test assessment criteria.