An Efficient Machine Learning Methodology for Liver Computerized Tomography Image Analysis

A disease that can decrease the performance of a liver and influence the execution method of hormone, protein, and nutrients in the human body is known as liver disease. This liver segmentation will be a very useful technique in computer-based liver disease diagnosis as well as in surgery planning. The abdominal CT (computerized tomography) scan will be used to identify the liver disease. This scan technique will generate images of organs with better quality and accuracy than the conventional x-ray photo method. The Abdominal CT- Scan is unable to capture the images of the heart with a better resolution and accuracy. To generate liver locations in the segmentation method, an algorithm called watershed transform will be used. These locations are useful in determining the objects by background. The binary threshold technique will be used in further image segmentation, which will help to separate the image of the liver from the object. Finally, the percentage of the affected liver area will be determined by using calculations. The image quality of the abdominal CT scan
will be effectively enhanced by using an adaptive filter. A
deep learning method comprises the transfer function and
will be managed by changed parameters. Here all these
parameters will be adjusted based on the optimization
process. The liver affected or disease area will be used as
radiology in the analysis of doctors. So the watershed
method using an adaptive filter will be used for the detection of liver disease using an abdominal CT scan.