Bone Fracture Image Denoising using a Novel Filter with Hybrid Optimization

Objective: The prime objective of this particular research is to enhance the quality of Bone fracture medical images by noise removal in the preprocessing stage, ensuring much better detection and analysis of anomalies in the human body. At the same time, it presents a solution to some of the deficits of traditional digital filters with regard to the optimization robustness and time efficiency. Method: Introduces a Decimation Based Guided Box Filter novelty used in the first time for Bone fracture Medical Image Denoising. For performance optimization, a hybrid algorithm based on the Particle Swarm optimization with Ant Colony Optimization (PSOACO) is adopted. The hybrid method ensures that the selection of optimal parameters for the filter is across various noise types. Findings: The experimental PSOACO model proved to be superior performance in removing any of the four considered kinds of noise, since it gave the marked PSNR of ~38.6 dB, SSIM of ~0.99, and MSE of ~0.01, which are higher than those produced by PSO, CS, or FF algorithms in any way. The method was found to be effective in terms of perceived quality and numerical measures alike. Novelty: This study reports on the first-time use of a decimation box filter specifically aimed at medical image denoising, combining this innovative filter together with a hybrid bio-inspired optimization method for higher perform ability. The approach is thus highly innovative with respect to both the filtering technique as well as the optimization strategies. Application: The method can be used in any medical imaging systems, especially in diagnostic and pre-processing tools, and it can also be integrated into AI platforms for an improved image interpretation and clinical accuracy. Other possibilities of applications include tumor detection, organ segmentation, and automated diagnostic support. Keywords: Digital filters, Bone Fracture medical images, PSO, ACO, FF, Bio-inspired optimization