Chimp Optimization Algorithm Influenced Type-2 Intuitionistic Fuzzy C-Means Clustering-Based Breast Cancer Detection...

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Computational Biology

Background:

• Early breast cancer detection is crucial for effective treatment and improved patient outcomes.
• Accurate image analysis is vital for identifying malignancy at its earliest stages.
• Existing clustering methods for medical image analysis have limitations in accuracy and efficiency.

Purpose of the Study:

• To propose a novel algorithm, Chimp Optimization Algorithm Based Type-2 Intuitionistic Fuzzy C-Means Clustering (COA-T2FCM), for highly accurate breast cancer detection.
• To enhance the precision of malignancy detection through an optimized clustering approach.
• To evaluate the performance of the proposed COA-T2FCM algorithm against conventional methods.

Main Methods:

• Developed the COA-T2FCM algorithm by integrating type-2 intuitionistic fuzzy c-means clustering with an oppositional function.
• Utilized the Chimp Optimization Algorithm to optimize cluster centers and fuzzifiers within the clustering process.
• Assessed performance using metrics like accuracy, specificity, sensitivity, Jaccard Similarity Index (JSI), and Dice Similarity Coefficient (DSC) on mammogram datasets (Mini-MIAS, DDSM, Inbreast).

Main Results:

• The proposed COA-T2FCM method achieved an average accuracy of 97.29% and a Jaccard Index Score (JSI) of 95%.
• Demonstrated superior performance compared to traditional Fuzzy C-Means and K-Means clustering techniques.
• Validation on multiple public mammogram datasets confirmed the algorithm's effectiveness and accuracy in identifying cancerous regions.

Conclusions:

• The COA-T2FCM algorithm offers a significant advancement in automated breast cancer detection from mammograms.
• The proposed method provides high accuracy and reliability, crucial for early-stage diagnosis.
• COA-T2FCM shows strong potential for clinical application in improving breast cancer screening and diagnosis.