Related Experiment Video
Updated: Jul 19, 2025
04:48
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
2.8K
Classification of Histopathological Images from Breast Cancer Patients Using Deep Learning: A Comparative Analysis
Louie Antony Thalakottor1, Rudresh Deepak Shirwaikar2, Pavan Teja Pothamsetti1
1Department of Information Science and Engineering, Ramaiah Institute of Technology (RIT), 560054, India.
Critical Reviews in Biomedical Engineering
|August 15, 2023
Summary
Deep learning models can automate breast cancer detection from histopathology images, improving diagnostic speed and accuracy. The DenseNet201 model achieved 91.3% accuracy, showing promise for replacing manual analysis.
Area of Science:
- Oncology
- Digital Pathology
- Artificial Intelligence
Background:
- Early cancer detection significantly improves survival rates.
- Manual analysis of histopathology slides for breast cancer biopsy is time-consuming.
- Automated techniques show potential for faster and more accurate abnormality detection.
Approach:
- Utilized the breast cancer histopathological database (BreakHis) for image analysis.
- Applied three convolutional neural network (CNN) models: VGG19, DenseNet201, and ResNet50V2.
- Processed images to enhance features, classified them, and evaluated algorithm accuracy.
Key Points:
- Compared the performance of VGG19, DenseNet201, and ResNet50V2 CNN models.
- DenseNet201 demonstrated superior performance among the tested models.
- Achieved a diagnostic accuracy of 91.3% using the DenseNet201 model.
Conclusions:
- Deep learning algorithms, particularly CNNs, show significant potential for automating breast cancer diagnosis.
- The DenseNet201 model offers a promising alternative to manual histopathological analysis.
- Further development of AI-based methods could revolutionize breast cancer detection and patient outcomes.
