Lung Cancer: the Chemotherapy Success Formula

Essay details

Please note! This essay has been submitted by a student.

Download PDF

Fundamentally any lung cancer detection system uses a variety of algorithms and techniques for pre-processing, image segmentation and feature extraction which lead to a difference in accuracy and results. Such methods and some of their details are mentioned in this section.

A. Kulkarni proposed a system for lung nodule cancer detection using CT images in DICOM format. Image smoothing was done by Median filter to reduce blurring of edges. The advantage of using a median filter is that it is not affected by individual noise spike, to eliminate impulsive noise quite well and it does not blur edges much and can be applied iteratively.

Essay due? We'll write it for you!

Any subject

Min. 3-hour delivery

Pay if satisfied

Get your price

Gabor filter is used for enhancement as it gives a better result compared to Fast Fourier Transform and auto enhancement. Presentation of the image based on Gabor function represents an excellent local and multi-scale decomposition that is simultaneously localization in space and frequency domain. Marker controlled Watershed algorithm is used for segmentation purpose. Area and parameter was extracted feature based on which classification was done.The median filter gives more accurate results compared to Gaussian, mean and Wiener filters.

K. Punithavathy et al. presented a methodology for automatic lung cancer detection in PET/CT images. Along with Wiener filtering, for pre-processing contrast level, adaptive histogram equalization (CLAHE) technique is also used. Morphological operations like closing and opening are performed for accurate extraction of lung ROI. Feature classification is done using fuzzy clustering method. FCM is unsupervised, simpler and soft clustering method that retains more information of the image as compared to hard clustering method.

These Morphological operations enable accurate lung ROI extraction and reduce the search space.

N. Hadavi suggested a technique Gabor filter is used for performing Image enhancement. Due to its advantages such as fast processing and easy influence, thresholding technique can be used for image segmentation. Features like nodule size, shape, contrast and the region for analysis are extracted.{Cellular Learning Automata (CLA)} model is obtained by developing the cellular automata including a learning automaton to each cell. CLA model is designed for components according to experiences of themselves and other components experiences are trained and CLA also have the capability to improve their behavior.Cellular learning automata are well trained; the model is capable of reduced rate of error and hence enhances the system’s reliability.

S. Kanitkar introduced a novel approach to detect a lung cancer using image processing technique. The Gaussian filter is used to smooth the image in the preprocessing stage so that it can remove high-frequency components from the image transform is used for the segmentation purpose. The features such as average intensity, perimeter, and area are extracted from the detected tumor. To extract the region minimum value from image watershed segmentation is used. It determines to the divide a line with the least value. The dividing line in a form of the image can give the rapid change of boundary. It behaves the image as a plane, where light pixels are high and dark pixels are low.The marker-controlled watershed segmentation technique separates the touching objects in the image.It provides the best identification of the main edge of the image and also avoids over-segmentation.

B. Patil proposed an approach to predict the probability of lung cancer detection. Binarization is the initial approach is and the second approach is masking. The images that are used in the analysis are in standard JPEG format. For converting them in Grey level ‘Otsu’s’ method can be used. Marker-controlled watershed segmentation is used that gives an accuracy of 85.27%. For prediction of lung cancer using Binarization approach dependent on a number of black pixels is greater than white pixels in normal lung images is used. So the counting starts from the black pixels for normal and abnormal images, if the number of the black pixels of a new image is greater than the threshold, then it are indicated that the image is normal, otherwise, vice versa. Another method for prediction is masking that depends on the fact that the masses are appeared as white linked areas inside ROI, as they increase the percent of cancer presence increase.

Initially, Haralick described that the and its attributes are tools for image classification. The grey level co-occurrence matrix (GLCM) was initially designed for texture classification of two-dimensional images. To calculate GLCM attributes for three-dimensional data it is necessary to adapt the methodology to work in 3D space. Therefore, it is important for the classified data to adapt the GLCM calculation to work in the three-dimensional space.

C. T. Henschke proposed an automatic computer-aided diagnosing system for identifying lung cancer by analyzing lung CT images. Wiener filter, erosion, slicing techniques was used by authors to extract lung image region from CT image. Bit image slicing was used to convert CT images to binary image region growing segmentation. The algorithm is used to segment extracted lung regions. After segmentation of the lung region rule-based model was used to classify the cancer nodules.

Get quality help now

Dr. Diane

Verified writer

Proficient in: Other Diseases & Conditions, Oncology

4.9 (280 reviews)
“She understood my main topic well and follow the instruction accordingly. She finished the paper in a timely manner! I would definitely hire her again! ”

+75 relevant experts are online

banner clock
Clock is ticking and inspiration doesn't come?
We`ll do boring work for you. No plagiarism guarantee. Deadline from 3 hours.

We use cookies to offer you the best experience. By continuing, we’ll assume you agree with our Cookies policy.