Operational Research and Reconstruction Methods in Medical Imaging

With the widespread availability of 3D printing technology, there's potential to address the issue of printing replacement parts for broken objects. Traditional methods of 3D Printing will face a Challenge to replicate accurately for broken pieces, specifically for the fracture lines that can be complex for geometry. In this paper we discuss about a novel approach: Neural Network system which is optimized in Hybrid is designed for reconstructing 3D objects automatically such as toys, vessels, pots and medical related images. This process includes several key stages like acquisition of image, preprocessing, extraction of features, recognition, alignment and matching of fragments. First to eliminate noise from objects they are scanned by using preprocessing so that the data is clean for the input and forwarded for next step. To identify and quantify geometric feature we use extraction of feature, texture characteristics, fragments of boundaries and edges are also extracted. In next stage we try to determine placement of fragments that are broken within the object in the system to match accordingly. A function of fitness which is hybrid uses techniques like optimization to align and fit for fragments to improve accuracy.