Related Experiment Video
Updated: Apr 20, 2026
05:33
Introduction of an Integrated Pathology Image Management, Artificial Intelligence, and Reporting System
Published on: July 11, 2025
1.5K
Computer-aided diagnosis system based on fuzzy logic for breast cancer categorization
Gisele Helena Barboni Miranda1, Joaquim Cezar Felipe1
1Department of Computing and Mathematics, Faculty of Philosophy, Sciences and Languages of Ribeirão Preto, University of São Paulo at Ribeirão Preto, Avenida Bandeirantes, 3900, Ribeirão Preto 14040-901, SP, Brazil.
Computers in Biology and Medicine
|December 3, 2014
Summary
This study introduces a fuzzy logic method for automated breast lesion classification, improving semantic consistency in medical image analysis. The system achieved 76.67% accuracy for nodules and 83.34% for calcifications.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Fuzzy Logic
Background:
- Fuzzy logic addresses challenges in computational systems simulating radiologist reasoning for medical image analysis.
- A novel method enhances semantic consistency in image description features.
- A computer-aided diagnosis tool for automatic Breast Imaging Reporting and Data System (BI-RADS) categorization of breast lesions was developed.
Purpose of the Study:
- To apply fuzzy logic concepts for improved representation of image features in medical analysis.
- To develop an automated tool for BI-RADS categorization of breast lesions.
- To enhance the semantic consistency and reliability of computer-aided diagnosis systems.
Main Methods:
- Defined malignancy values for image descriptors based on BI-RADS standards.
- Developed a fuzzy inference system using the Fuzzy Omega algorithm for membership function generation.
- Utilized statistical analysis of datasets to map class distributions for algorithm training.
Main Results:
- Physician evaluations were analyzed using the Fuzzy Omega algorithm.
- Achieved an accuracy of 76.67% for nodule classification.
- Achieved an accuracy of 83.34% for calcification classification.
Conclusions:
- The method integrates linguistic rules and numerical models for better specialist-computer interaction.
- This approach leads to more effective and reliable results in medical image analysis.
- The developed system shows potential for improving diagnostic accuracy in breast lesion categorization.
