Purpose:To develop a system to facilitate the retrieval of radiologic images that contain similar-appearing lesions and to perform a preliminary evaluation of this system with a database of computed tomographic (CT) images of the liver and an external standard of image similarity. Materials and Methods:Institutional review board approval was obtained for retrospective analysis of deidentifi ed patient images. Thereafter, 30 portal venous phase CT images of the liver exhibiting one of three types of liver lesions (13 cysts, seven hemangiomas, 10 metastases) were selected. A radiologist used a controlled lexicon and a tool developed for complete and standardized description of lesions to identify and annotate each lesion with semantic features. In addition, this software automatically computed image features on the basis of image texture and boundary sharpness. Semantic and computer-generated features were weighted and combined into a feature vector representing each image. An independent reference standard was created for pairwise image similarity. This was used in a leaveone-out cross-validation to train weights that optimized the rankings of images in the database in terms of similarity to query images. Performance was evaluated by using precisionrecall curves and normalized discounted cumulative gain (NDCG), a common measure for the usefulness of information retrieval. Results:When used individually, groups of semantic, texture, and boundary features resulted in various levels of performance in retrieving relevant lesions. However, combining all features produced the best overall results. Mean precision was greater than 90% at all values of recall, and mean, best, and worst case retrieval accuracy was greater than 95%, 100%, and greater than 78%, respectively, with NDCG. Conclusion:Preliminary assessment of this approach shows excellent retrieval results for three types of liver lesions visible on portal venous CT images, warranting continued development and validation in a larger and more comprehensive database.q RSNA, 2010

No abstract

No abstract

No abstract

The research of application models based on traditional convolutional neural networks has gradually entered the bottleneck period of performance improvement, and the improvement of chest X-ray image models has gradually become a difficult problem in the study. In this paper, the Swin Transformer is introduced into the application model of pneumonia recognition in chest X-ray images, and it is optimized according to the characteristics of chest X-ray images. The experimental results based on the model in this paper are compared with those of the model built with the traditional convolutional neural network as the backbone network, and the accuracy of the model is proved to be greatly improved. After the comparison experiments on two different datasets, the experimental results show that the accuracy of the model in this paper improves from 76.3% to 87.3% and from 92.8% to 97.2%, respectively. The experiments show that the accuracy of image enhancement based on the features of chest X-ray images in this model will be higher than the accuracy without image enhancement. In the experiments of this paper, the identification decision factors in the chest X-ray images were extracted by grad-cam combined with a transformer to find the corresponding approximate lesion regions.