Related Experiment Video
Updated: Aug 3, 2025
13:44
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
Published on: August 30, 2013
43.0K
A Stochastic Grammar Approach to Mass Classification in Mammograms
Summary
Syntactic approaches using stochastic grammars effectively classify breast masses from mammograms. This method achieves high accuracy (96-100%) even with limited training data, outperforming other machine learning techniques.
Area of Science:
- Medical Imaging
- Computational Biology
- Machine Learning
Background:
- Breast cancer is a leading cause of cancer deaths in women globally.
- Early and accurate diagnosis significantly improves survival rates.
- Traditional machine learning methods for breast cancer diagnosis often require extensive training datasets.
Purpose of the Study:
- To introduce and evaluate a syntactic approach for classifying mammographic masses as benign or malignant.
- To demonstrate the efficacy of grammar-based methods in breast cancer diagnosis, particularly with limited data.
Main Methods:
- Features were extracted from polygonal representations of mammographic masses.
- A stochastic grammar approach was employed for mass classification.
- Performance was evaluated against other machine learning techniques.
Main Results:
- The grammar-based classifiers achieved superior performance in mass classification.
- Accuracies ranged from 96% to 100%, demonstrating robustness.
- The approach effectively discriminated between benign and malignant masses using small image samples.
Conclusions:
- Syntactic approaches are robust and effective for breast mass classification, even with limited training data.
- Grammar-based methods show potential for wider application in mammogram analysis.
- This technique offers a promising alternative to traditional machine learning models requiring large datasets.
