The Development of a Fuzzy Logic System Using MATLAB for Early Detection of Hereditary Cancer in BRCA1/2 Negative...

Area of Science:

• Medical Informatics
• Computational Biology
• Genetics

Background:

• Hereditary breast cancer (BC) poses a significant health challenge.
• Accurate and early detection of BC, especially in BRCA1/2-negative cases, is crucial for effective treatment.
• Existing diagnostic tools may benefit from enhanced computational approaches for genetic variant interpretation.

Purpose of the Study:

• To develop and validate an artificial intelligence (AI) software tool for the rapid detection of hereditary breast cancer (BC).
• To utilize a fuzzy logic system in MATLAB to analyze genetic variants associated with BC in individuals with negative BRCA1/2 genes.
• To create a clinical decision-support system integrating clinical and genetic data for early BC classification.

Main Methods:

• Development of a fuzzy logic system using MATLAB to analyze hereditary BC risk factors and gene mutations.
• Integration of clinical and genetic data from 90 relevant patients (out of 488 studied) with negative BRCA1/2 mutations.
• Training the system on 90 cases and validating its accuracy on six independent patients.

Main Results:

• The AI system demonstrated reliable accuracy in assessing genetic variants.
• Identified pathogenic variants with 92% probability, benign variants with 25% probability, and variants of unknown significance with 50% probability.
• The system successfully classified genetic variants based on integrated clinical and genetic data.

Conclusions:

• The developed AI software shows promise as a clinical decision-support tool for early breast cancer detection.
• Fuzzy logic systems can be effectively applied to interpret complex genetic data for BC risk assessment.
• Further research and clinical application of this AI tool could significantly advance hereditary breast cancer diagnosis.