Deep Neural Framewrork for Early and Accurate Breast Cancer Diagnosis
- Version
- Download 9
- File Size 673.98 KB
- File Count 1
- Create Date 22 March 2026
- Last Updated 22 March 2026
Deep Neural Framewrork for Early and Accurate Breast Cancer Diagnosis
V.Tharani
Dept of Computer Science and Engineering Dr.M.G.R Educational and Research Institute Chennai,India
tharaniv127@gmail.com
Dr.M.Manikandan
Dept of Computer Science and Engineering Dr.M.G.R.Educational and Research Institute Chennai,India
manikandan2602@gmail.com
U.Anushiya
Dept of Computer Scienceand Engineering Dr.M.G.R.Educational and Research Institute Chennai,India
anushiyau@gmail.com
Dr.V.Sai Shanmuga Raja
Dept of Computer Science and Engineering Dr.M.G.R.Educational andResearch Institute Chennai,India
saishanmugaraja.cse@drmgrdu.ac.in
S.Priya
Dept of Computer Science and EngineeringDr.M.G.R.Educational and Research Institute Che nnai,India
priyaselvakumar495@gmail.co m
Dr.M.Sujitha
Dept of Computer Science and Engineering Dr.M.G.R Educational andResearch Institute Chennai,India
sujitha.ece@drmgrdu.ac.in
Abstract— Breast Cancer is the most common disease which as diagnosed malignant tumor in women in the worldwide. This disease crucial need for early and accurate detection. The existing diagnostic system approaches such as manual interpretation of mammograms are time consuming and prone human error. The primary aim of this project is to utilize deep neural framework for early and accurate breast cancer using CBIS-DDSM datasets. This system employs an EfficientNet- B4,ResNeXt , and Swin Transformer architectures to classify the breast tumor as malignant or benign or malignant.Thissystems image preprocessing techniques including resizing,normalization, and Contrast Limited Adaptive Histogram Equalization(CLAHE) are employed to enhance image quality of datasets.The system also have explainable artificial intelligence byusing Gradient-Weighted Class Activation Mapping to provide a visual result of model prediction with 98.5% of accuracy. This research contributes to advancing early and accurate breast cancer detection through deep learning, promising improved outcomes.Keywords— Breast Cancer Detection, Deep Learning, Efficient NetB4, ResNeXt , Swin Transfromer,Grad-CAM.
Download