Efficient Breast Cancer Diagnosis from Complex Mammographic Images Using Deep Convolutional Neural Network

Area of Science:

• Medical imaging
• Artificial intelligence in healthcare
• Oncology

Background:

• Breast cancer is a leading cause of death in women, necessitating early and accurate detection.
• Digital mammography is crucial for breast cancer diagnosis, but early-stage detection remains challenging.
• Computer-aided diagnosis (CAD) systems enhance radiologists' ability to detect lesions.

Purpose of the Study:

• To develop and evaluate a computational framework for breast cancer diagnosis using deep convolutional neural networks.
• To classify mammogram images as benign or malignant using the ResNet-50 architecture.
• To improve the accuracy and efficiency of early breast cancer detection.

Main Methods:

• Utilized a ResNet-50 convolutional neural network (CNN) architecture.
• Employed transfer learning, pretraining the ResNet-50 model on the ImageNet dataset.
• Trained and classified the INbreast dataset, categorizing images into benign and malignant classes.

Main Results:

• The proposed framework achieved a classification accuracy of 93% on the INbreast dataset.
• The model demonstrated superior performance compared to other models trained on the same dataset.
• The deep learning approach showed high accuracy in classifying various mammograms.

Conclusions:

• The developed framework facilitates early diagnosis and classification of breast cancer, distinguishing between benign and malignant tumors.
• Deep convolutional neural network algorithms can achieve highly accurate results in mammogram analysis.
• This approach has the potential to enhance medical diagnostic tools by reducing screening error rates and saving lives.