Related Experiment Video
Updated: May 29, 2025
03:31
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
Published on: December 15, 2023
455
An Efficient Lightweight Multi Head Attention Gannet Convolutional Neural Network Based Mammograms Classification
Ramkumar Muthukrishnan1, Ashok Balasubramaniam2, Vijaipriya Krishnasamy3
1Electronics and Communication Engineering, Sri Krishna College of Engineering and Technology, Coimbatore, India.
Summary
Deep learning enhances breast cancer detection using a novel Lightweight Multihead attention Gannet Convolutional Neural Network (LMGCNN). This automated system achieves high accuracy in classifying mammograms, improving early diagnosis and patient outcomes.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Oncology
Background:
- Automated breast cancer detection in mammograms is crucial for early diagnosis.
- Challenges include time constraints, feature extraction difficulties, and limited training data.
- Deep learning offers a promising solution to these limitations.
Purpose of the Study:
- To develop an automated deep learning system for improved breast cancer detection and classification in mammograms.
- To address the challenges faced by medical professionals in mammogram analysis.
Main Methods:
- Introduction of a Lightweight Multihead attention Gannet Convolutional Neural Network (LMGCNN) for mammogram classification.
- Image enhancement techniques including wiener filtering, unsharp masking, and adaptive histogram equalization.
- Feature extraction using Grey-Level Co-occurrence Matrix (GLCM) and optimized selection via a self-adaptive quantum equilibrium optimizer with artificial bee colony.
Main Results:
- The LMGCNN model was evaluated on the CBIS-DDSM and MIAS datasets.
- Achieved high accuracy rates of 98.2% and 99.9% on the respective datasets.
- Demonstrated superior performance in breast cancer detection compared to existing models.
Conclusions:
- The developed method shows significant potential for assisting in early and accurate breast cancer detection.
- This advancement could lead to improved patient outcomes through timely diagnosis and treatment.
