Please use this identifier to cite or link to this item: http://itr.iub.edu.pk:8000/xmlui/handle/123456789/1517
Title: Lungs Nodule Cancer Detection using CT scan images
Authors: Abbas, Waseem
Keywords: Electronic Engineering
Issue Date: 2020
Series/Report no.: ;5707
Abstract: In the lungs, the formation of nodules is like a globular lesion. At initial stages, it may be non-cancerous, but over time the nodule turned into the cancerous part. Early detection at the beginning stage of the formation of lung nodules is important for the patient to get the chance of survival from the disease. However, the detection of a nodule in the early stages is a challenging problem. The detection of lung nodule cancer by Computer-Aided Diagnosis (CAD) systems is a great support to the medical experts and lowers the death rate. In this thesis, the CAD system is established for the early detection of lung nodule cancer. Firstly, the input image is converted into grayscale. Contrast Limited Histogram Equalization (CLAHE) is utilized on grayscale image for contrast enhancement of the image. Then, Otsu thresholding is used for segmentation. The background and other geometrical objects are removed by using morphological filters and to extract the nodule. The resultant image is de-noised by applying a Discrete Wavelet Transform (DWT). Grey Level Co-occurrence Matrix (GLCM) is used for the extraction of features such as correlation, energy, etc. along with the application of Principle Component Analysis (PCA) for feature selection. To classify the image into benign (non-cancerous) or malignant (cancerous) Support Vector Machine (SVM) is used. The performance parameters such as accuracy, sensitivity, specificity, Peak Signal to Noise Ratio (PSNR), Root Mean Square Errors (RMSE), and Area under the Curve are used for the evaluation of the proposed method. The proposed system is verified on the LIDC dataset and produced better results as compared to other existing methods. The visual and parametric results of the proposed method are computed and compared.
URI: http://itr.iub.edu.pk:8000/xmlui/handle/123456789/1517
Appears in Collections:Thesis

Files in This Item:
File Description SizeFormat 
5707 Lungs Nodule Cancer Detection using.pdf386.77 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.