No abstract

Around the world, more than 250 countries are affected by the COVID-19 pandemic, which is caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). This outbreak can be controlled only by the diagnosis of the COVID-19 infection in early stages. It is found that the radiographic images are ideal for the fastest diagnosis of COVID-19 infection. This paper proposes an ensemble model which detects the COVID-19 infection in the early stage with the use of chest X-ray images. The transfer learning enables to reuse the pretrained models. The ensemble learning integrates various transfer learning models, i.e., EfficientNet, GoogLeNet, and XceptionNet, to design the proposed model. These models can categorize patients as COVID-19 (+), pneumonia (+), tuberculosis (+), or healthy. The proposed model enhances the classifier's generalization ability for both binary and multiclass COVID-19 datasets. Two popular datasets are used to evaluate the performance of the proposed ensemble model. The comparative analysis validates that the proposed model outperforms the state-of-art models in terms of various performance metrics.

No abstract

Content-based image retrieval (CBIR) system generally retrieves images based on the matching of the query image from all the images of the database. This exhaustive matching and searching slow down the image retrieval process. In this paper, a fast and effective CBIR system is proposed which uses supervised learning-based image management and retrieval techniques. It utilizes machine learning approaches as a prior step for speeding up image retrieval in the large database. For the implementation of this, first, we extract statistical moments and the orthogonal-combination of local binary patterns (OC-LBP)-based computationally light weighted color and texture features. Further, using some ground truth annotation of images, we have trained the multi-class support vector machine (SVM) classifier. This classifier works as a manager and categorizes the remaining images into different libraries. However, at the query time, the same features are extracted and fed to the SVM classifier. SVM detects the class of query and searching is narrowed down to the corresponding library. This supervised model with weighted Euclidean Distance (ED) filters out maximum irrelevant images and speeds up the searching time. This work is evaluated and compared with the conventional model of the CBIR system on two benchmark databases, and it is found that the proposed work is significantly encouraging in terms of retrieval accuracy and response time for the same set of used features.

Artificial intelligence (AI) has made various developments in the image segmentation techniques in the field of medical imaging. This article presents a liver tumor CT image segmentation method based on AI medical imaging-based technology. This study proposed an artificial intelligence-based K-means clustering (KMC) algorithm which is further compared with the region growing (RG) method. In this study, 120 patients with liver tumors in the Post Graduate Institute of Medical Education & Research Hospital, Chandigarh, India, were selected as the research objects, and they were classified according to liver function (Child–Pugh), with 58 cases in grade A and 62 cases in grade B. The experimentation indicates that liver tumor showed low density on plain CT scan, moderate enhancement in the arterial phase of the enhanced scan, and low-density filling defect in the involved blood vessel in the portal venous phase (PVP). It was observed that the CT examination is more sensitive to liver metastasis than hepatocellular carcinoma ( P < 0.05 ). The outcomes obtained depict the good deposition effect of lipiodol chemotherapy emulsion (LCTE) in the contrast group with rich blood type accounted for 53.14% and the patients with the poor blood type accounted for 25.73% showed poor deposition effect. The comparison with the state-of-the-art method reveals that the segmentation effect of the KMC algorithm is better than that of the conventional RG method.