Related Experiment Video
Updated: Jul 25, 2026
07:15
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
A Novel Tet-LDN Based Feature Extraction and Parallel Convolutional LeNet for Breast Cancer Detection and
Eliganti Ramalakshmi1, Loshma Gunisetti2, Sumalatha Lingamgunta3
1Research Scholar, JNTUK, Department of Information Technology, Chaitanya Bharathi Institute of Technology, Gandipet, Rangareddy, Telangana, 500075, India.
Cancer Investigation
|May 28, 2026
Summary
This study introduces a new Parallel Convolutional-LeNet (PConv-LeNet) method for breast cancer (BC) detection and classification using histopathological images. The PConv-LeNet achieved high accuracy, demonstrating its potential for improved BC diagnosis.
Area of Science:
- Medical Imaging
- Computational Biology
- Artificial Intelligence in Medicine
Background:
- Breast cancer (BC) remains a leading cause of mortality globally, necessitating advanced diagnostic tools.
- Histopathological image analysis is crucial for accurate BC detection and classification.
- Existing methods may face challenges in noise reduction, feature extraction, and classification accuracy.
Purpose of the Study:
- To propose a novel Parallel Convolutional-LeNet (PConv-LeNet) technique for enhanced BC detection and classification.
- To improve the accuracy and reliability of BC diagnosis using histopathological images.
- To develop an automated system for BC analysis.
Main Methods:
- Utilized histopathological images for BC analysis.
- Applied Medav Filter for image denoising and Parallel Reverse Attention Network (PraNet) for blood cell segmentation.
- Performed feature extraction using shape features and the proposed Tetrolet-Local Direction Number (Tet-LDN).
- Employed a Parallel Convolutional Neural Network (PCNN) tuned by Remora Kill Herd Optimization (RKHO) for BC detection.
- Developed and utilized the PConv-LeNet, integrating PCNN and LeNet, for BC classification.
Main Results:
- The PConv-LeNet model achieved 91.78% accuracy.
- The model demonstrated 91.88% specificity in BC detection.
- A sensitivity of 92.38% was attained by the PConv-LeNet for BC classification.
Conclusions:
- The proposed PConv-LeNet technique shows significant promise for accurate breast cancer detection and classification.
- The integration of advanced image processing and deep learning models enhances diagnostic capabilities.
- This approach offers a potential improvement in the early and reliable diagnosis of breast cancer.