Please use this identifier to cite or link to this item:
http://hdl.handle.net/10603/545839
Title: | An efficient approach for the detection and classification of cervical cancer |
Researcher: | Subarna T G |
Guide(s): | Sukumar P |
Keywords: | Computer Tomography Convolutional Neural Networks Tumor Detection Accuracy |
University: | Anna University |
Completed Date: | 2023 |
Abstract: | Machine and Deep learning algorithms have many significant newlineadvantages which is used in computer vision based automatic applications, newlinepattern identification and health monitoring based automation applications. In newlinecomputer vision based applications, it is widely used nowadays in medical newlineimage processing which is used for analyzing and segmenting different patterns newlinefor different modality of images such as Computer Tomography (CT) and newlineMagnetic Resonance Imaging and ultra Sound (US). Due to these advantages newlineof machine and deep learning algorithms, this research work uses novel newlinemethodology to detect and classify the cervical images into either normal or newlineabnormal. newlineIn proposed cervical cancer detection method-1, the cancer regions newlinein cervigram images are segmented using machine learning classifier. The newlineGabor transform transforms the pixel coordinates of the source cervigram newlineimage into multi orientation coordinates and then the systematic feature newlineproperties are computed from the transformed cervigram image. These newlinecomputational systematic features are further classified through the Adaptive newlineNeuro Fuzzy Inference System (ANFIS) classification process, which produces newlinethe classification results of either Abnormal case cervical image or Normal newlinecase cervical image .TheMean Index Rate (MIR) for Guanacaste dataset newlinecervical images is about 91.25% and the average MIR for MCSI dataset newlinecervical images is about 77.7%. The proposed method stated in this method-1 newlineconsumed 2.1 ms detection time for processing of each cervical image in newlineGuanacaste dataset and also consumed 2.8 ms detection time for processing of newlineeach cervical image in MSCI dataset. In this research work, Intel Pentium Core newlinei7 processor, 1 TB harddisk, 8 GB RAM hardware units are used to measure newlinethe detection time. newline |
Pagination: | xviii,122p. |
URI: | http://hdl.handle.net/10603/545839 |
Appears in Departments: | Faculty of Information and Communication Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
01_title.pdf | Attached File | 3.6 MB | Adobe PDF | View/Open |
02_prelimpage.pdf | 4.84 MB | Adobe PDF | View/Open | |
03_contents.pdf | 3.59 MB | Adobe PDF | View/Open | |
04_abstracts.pdf | 3.6 MB | Adobe PDF | View/Open | |
05_chapter1.pdf | 3.59 MB | Adobe PDF | View/Open | |
06_chapter2.pdf | 3.59 MB | Adobe PDF | View/Open | |
07_chapter3.pdf | 3.57 MB | Adobe PDF | View/Open | |
08_chapter4.pdf | 3.58 MB | Adobe PDF | View/Open | |
09_chapter5.pdf | 3.56 MB | Adobe PDF | View/Open | |
10_annexure.pdf | 117.9 kB | Adobe PDF | View/Open | |
80_recommendation.pdf | 97.98 kB | Adobe PDF | View/Open |
Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Altmetric Badge: