An approach to predict gallbladder stone using Mumford Shah approach algorithm and comparing accuracy with K-means clustering algorithm

• Main Authors: Babuji, D., Manikandan, G.
• Format: Conference Proceeding
• Language: English
• Subjects: Algorithms; Cluster analysis; Clustering; Datasets; Gallbladder; Performance evaluation; Stone; Vector quantization
• ISSN: 0094-243X; 1551-7616
• DOI: 10.1063/5.0233967

The purpose of this research is to develop a novel approach to the prediction of gallbladder stones by utilising the Mumford Shah Approach. Then, in order to evaluate the effectiveness of this method, a comparison will be made between its performance and that of the K-means clustering algorithm. Methodologies and instruments for research: This research made use of a dataset that was made accessible to the public on Kaggle. The dataset contained 5,350 photographs taken by 726 patients. For the purpose of evaluation, twenty photos were provided to each of the two groups: one group utilised the Novel Mumford Shah approach, and the other group utilised the Novel K-means clustering algorithm. An alpha of 0.05, a beta of 0.2, and a pretest power of 80 percent were utilised in the calculation of the sample size in order to determine the appropriate size of the sample. When it comes to the accuracy rate obtained from MATLAB simulation, the Novel Mumford Shah technique achieved 92.43 percent, while the Novel K-means clustering approach achieved 94.16 percent. From the findings, it can be observed that the Novel Mumford Shah Approach and K-means clustering exhibit statistically significant differences, as evidenced by a p-value of 0.001 (independent sample t-test p