Related Experiment Video
Updated: Aug 5, 2026
13:44
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
Published on: August 30, 2013
PWidFHNet: Parallel Wider Forward Harmonic Net to Classify Breast Cancer Using Histopathological Images
Rajesh Prasad1, Jayashree Prasad2, Nihar Ranjan3
1Department of Computer Science and Engineering, College of Engineering, Bharati Vidyapeeth Deemed to be University, Pune, India.
Cancer Biotherapy & Radiopharmaceuticals
|July 30, 2026
Summary
Accurate breast cancer classification is crucial for timely treatment. The novel PWidFHNet model, utilizing advanced image analysis and deep learning, achieves high accuracy in classifying breast cancer from histopathological images, aiding diagnostic decisions.
Area of Science:
- Oncology
- Medical Imaging
- Computer Science
Background:
- Breast cancer is a major global health concern for women.
- Accurate and timely diagnosis is vital for improving survival rates and reducing mortality.
- Current classification methods may lack the precision needed for effective treatment planning.
Purpose of the Study:
- To introduce the PWidFHNet model for precise breast cancer classification.
- To enhance the accuracy of breast cancer diagnosis from histopathological images.
Main Methods:
- Histopathological images undergo preprocessing with a Gaussian filter.
- Blood cell segmentation is performed using a Tversky-parallel reverse attention network.
- Feature extraction includes shape features and learned invariant feature transform.
- The PWidFHNet model, combining parallel convolutional neural networks and wide residual networks with harmonic analysis, performs the classification.
Main Results:
- The PWidFHNet model demonstrated high performance.
- Specificity reached 92.76%, accuracy 91.60%, and sensitivity 90.86%.
Conclusions:
- The PWidFHNet model offers a robust framework for breast cancer classification.
- It integrates advanced image processing, segmentation, feature extraction, and deep learning.
- The model shows potential as a diagnostic aid for pathologists in breast cancer assessment.