MULTI-MODAL MEDICAL IMAGE FUSION BASED ON MULTI-BASED BINARY RESIDUALFEATURE FUSION

Abstract

The term "medical image fusion" refers to the procedure of aligning and blending separate pictures captured using different imaging technologies. By enhancing imaging quality and decreasing unpredictability and redundancy, it aims to boost the clinical usefulness of medical pictures for the diagnosis and evaluation of medical conditions. Accuracy of clinical choices based on medical pictures has been enhanced via the use of multi model medical image fusion algorithms and tools. In this study, we show that combining pictures from chest X-rays and ultrasounds improve visibility. On the other hand, the merged picture is essential for a high-quality display. Improving the quality of the combined photos requires a classification technique. The accurate fusion and integration of medical images from multiple modalities is a critical step in the diagnostic process that leads to clinical activities and the appropriate treatment of patients, and this chapter suggests a Multi-based Binary Residual Feature Fusion (MBRFF) method for this purpose. The images are preprocessed usingthe Histogram equalization algorithm. For the feature extraction of medical images, the Multi- Threshold Contour Method is used. The shift variant decomposition transformation technique Discrete Wavelet Transform DWT is effective and it is used to decompose the medical image. The fused medical images are optimized by using Particle swarm optimization PSO. Finally, the simulation results reveal that the suggested mechanism outperforms the other techniques based on conventional approaches. Using the Origin Pro application, the study's outcomes are represented graphically