Home / Current Issue / Paper 1706647
Comparison of CAD Detection of Mammogram with SVM and CNN
Subject area: Science,Engineering and Technology · Area of research: Computer-Aided Detection
Abstract
This study compares the performance of Computer-Aided Detection (CAD) systems for mammogram analysis using two prominent machine learning techniques: These are the kind of models Support Vector Machines (SVM) and Convolutional Neural Networks (CNN). The main goal is, therefore, to assess the performance of these models in correctly classifying abnormal mammogram images, especially of early-stage breast cancer. The work uses a dataset of mammogram images with labels, removing outliers and repeating rows and columns, then normalizing equals and providing input data for both the SVM and CNN models. The former chosen quantities were introduced as performance metrics known as accuracy, precision, recall, and the area under the receiver operating characteristic curve (AUC) for both models. Research shows that the accuracy of both SVM and CNN is equivalent and that CNN has a higher sensitivity and specificity, indicating it could be more efficient in early cancer detection. The implications of these findings highlight the beneficial use of deep learning models in medical images, especially CNN models. This research is useful for the current development of CAD systems and gives potential future applications of AI in the context of diagnosis in clinics.
Keywords
Mammogram Detection, Breast Cancer, CAD Systems, SVM Models, CNN Models, Medical Imaging
How to cite this paper
@article{1706647,
author = {Mohit Jain, Arjun Srihari},
title = {Comparison of CAD Detection of Mammogram with SVM and CNN},
journal = {Iconic Research And Engineering Journals},
year = {2024},
volume = {8},
number = {6},
pages = {63-75},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1706647.pdf},
abstract = {This study compares the performance of Computer-Aided Detection (CAD) systems for mammogram analysis using two prominent machine learning techniques: These are the kind of models Support Vector Machines (SVM) and Convolutional Neural Networks (CNN). The main goal is, therefore, to assess the performance of these models in correctly classifying abnormal mammogram images, especially of early-stage breast cancer. The work uses a dataset of mammogram images with labels, removing outliers and repeating rows and columns, then normalizing equals and providing input data for both the SVM and CNN models. The former chosen quantities were introduced as performance metrics known as accuracy, precision, recall, and the area under the receiver operating characteristic curve (AUC) for both models. Research shows that the accuracy of both SVM and CNN is equivalent and that CNN has a higher sensitivity and specificity, indicating it could be more efficient in early cancer detection. The implications of these findings highlight the beneficial use of deep learning models in medical images, especially CNN models. This research is useful for the current development of CAD systems and gives potential future applications of AI in the context of diagnosis in clinics.},
keywords = {Mammogram Detection, Breast Cancer, CAD Systems, SVM Models, CNN Models, Medical Imaging},
month = {December},
}