Abstract:

Content-based Medical Image Retrieval (CBMIR) has been active in the research field over the past few years. A CBMIR system recovery's performance depends largely on some aspects, the nature of which has been studied in detail for decades. Although different strategies have been proposed, it is still one of the most challenging issues in current CBMIR research. This is mainly due to the so-called "semantic gap" between the device and the high-resolution image-low pixel capture. Today's medical images in daily practice have been increasingly created on the scale of millions. Recovering medical images from huge classifications is difficult, so it rises to the Content-Based Medical Image Retrieval System (CBMIR) system. In existing systems, images are first retrieved and then classified. Also, the retrieval rate obtained by the existing method is not ideal. Therefore in this work, a Clustering Segmentation and Multilevel Tetrolet Feature Extraction strategies are proposed to overcome the existing work drawbacks. The simulation of the proposed work is developed using Matlab simulation software. The predictive result for this work has encouraged performance, and the method is cost- effective compared to the purposes described in the literature. In terms of performance measurements, i.e., Accuracy, Precision and Recall.

Keywords: