Related Experiment Video
Updated: Aug 31, 2025
07:15
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
6.9K
A Deep Learning Computer-Aided Diagnosis Approach for Breast Cancer
Ahmed M Zaalouk1,2, Gamal A Ebrahim1, Hoda K Mohamed1
1Computer and Systems Engineering Department, Faculty of Engineering, Ain Shams University, Cairo 11517, Egypt.
Bioengineering (Basel, Switzerland)
|August 25, 2022
Summary
A deep learning computer-aided diagnosis (CAD) system was developed to aid pathologists in breast cancer diagnosis. The Xception model demonstrated superior performance, achieving high accuracies in classifying histopathological images.
Area of Science:
- Oncology
- Medical Imaging
- Artificial Intelligence
Background:
- Breast cancer poses a significant global health and economic burden, particularly affecting women.
- Histopathological examination of breast tissue biopsies is the current standard for diagnosis.
- Computer-aided diagnosis (CAD) systems can assist pathologists in improving diagnostic efficiency.
Purpose of the Study:
- To develop and evaluate a deep learning-based CAD system for breast cancer diagnosis using histopathological images.
- To compare the performance of five pre-trained convolutional neural network (CNN) models for breast cancer classification.
- To introduce a novel transfer learning approach for enhancing CAD system performance.
Main Methods:
- Five pre-trained CNN models (Xception, DenseNet201, InceptionResNetV2, VGG19, ResNet152) were analyzed.
- Data augmentation techniques were employed to improve model robustness.
- Transfer learning was utilized with histopathological images from the BreakHis dataset.
- Models were evaluated through magnification-dependent and independent binary and eight-class classifications.
Main Results:
- The Xception model exhibited the highest classification accuracies across all experimental conditions.
- Magnification-independent classification accuracies ranged from 93.32% to 98.99%.
- Magnification-dependent classification accuracies ranged from 90.22% to 100%.
Conclusions:
- The developed deep learning CAD system, particularly the Xception model, shows significant promise for accurate breast cancer diagnosis from histopathological images.
- The study highlights the effectiveness of deep learning and transfer learning in improving diagnostic capabilities in oncology.
- The findings suggest potential for integrating advanced AI tools into routine pathology workflows to aid in breast cancer detection.
