Development of cad system for CT lung cancer analysis and diagnosis using graphical user interface

Naha Azni, Naza Syahmie (2021) Development of cad system for CT lung cancer analysis and diagnosis using graphical user interface. Project Report. Melaka, Malaysia, Universiti Teknikal Malaysia Melaka. (Submitted)

	Text (24 Pages) Development of cad system for CT lung cancer analysis and diagnosis using graphical user interface.pdf - Submitted Version Download (462kB)
	Text (Full Text) Development of cad system for CT lung cancer analysis and diagnosis using graphical user interface.pdf - Submitted Version Restricted to Repository staff only Download (1MB)

Abstract

Development of a Computer Aided Design for Computed Tomography (CT) Lung Cancer Analysis and Diagnosis by displaying the picture in MATLAB utilising DICOM data from the LIDC dataset. CT is a non-invasive technique that produces CT pictures of any part of the human body without superimposition of neighbouring structures. The majority of lung tumour cells grow bigger and spread throughout the organs. The CT picture offers further information on the problem of lung cancer diagnosis by chest X-ray, however tiny tumours cannot be seen. To get the best possible result from the CT picture, two stages are required: pre-processing and segmentation. Pre-processing is used to isolate noise in CT images, while segmentation is used to divide the treated CT picture into several areas. The final result will be more accurate by pre-processing images, and the photos will be of more excellent quality. Thus, this work demonstrates how to construct a Graphical User Interface to display DICOM pictures using AppDesigner in MATLAB. To improve the quality of lung cancer segmentation, implying Otsu Threshold, it is possible to distinguish cancer that arises between other organs in the lung. The Otsu Threshold technique is a method for image thresholding based on clustering. It is effective if the histogram is bimodal. Essentially, the method seeks to decrease within-class variation while increasing between-class variance. The picture obtained by thresholding may be used to determine the cancer size's accuracy and compare it to the TCIA dataset with exact values.The ultimate goal of a wide variety of image processing applications is to extract significant characteristics from visual data for the computer to offer a description, interpretation, or understanding of the scene. The term "image processing" refers to the process of modifying or editing an existing image. Image processing aims to visually improve or statistically analyse some feature of the idea that is not immediately visible in its original state.

Actions (login required)

View Item