Clustering of Tuberculosis and Normal Lungs Based on Image Segmentation Results of Chan-Vese and Canny with K-Means

Abstract

The lungs are a vital organ in the human body. If there is interference with lung function, the health of the human body as a whole can be affected. Examination by medical workers needs to be done when there is interference with lung function. This examination can usually be done in various ways, one of which is through a chest X-ray radiographic examination procedure. The application of Artificial Intelligence is growing rapidly in the medical field, especially in diagnostics and treatment management. Artificial intelligence in the medical world can also be applied in processing image data in radiology to analyze X-ray results as supporting diagnostic information. Operators Chan-Vese and Canny are two edge detection operators in digital image processing in an effort to obtain the necessary information based on the shape and size of the object. This study was conducted for clustering of normal and tuberculosis lung conditions based on the results of chest X-ray image segmentation from Chan-Vese and Canny using K-Means Clustering. The results of clustering using K-Means obtained an accuracy value of 77.1%, a precision value of 88%, and a specificity value of 97.2%