Abstract: Early and accurate detection of breast cancer is essential for improving patient outcomes and tailoring treatment strategies. This study introduces a two-step machine learning framework for symptom-based breast cancer detection and carcinoma type identification. The initial step utilizes Random Forest (RF) to detect the presence of breast cancer based on extracted symptoms. If cancer is detected, the second step confirms the carcinoma type, specifically identifying ductal carcinoma, using Gray-Level Co-Occurrence Matrix (GLCM) for feature extraction and RF classification. The proposed system demonstrates enhanced accuracy and reliability, leveraging the strength of feature-based methods and ensemble learning techniques. This paper provides an in-depth analysis of methodologies, results, and related datasets, emphasizing the practicality and effectiveness of the system in clinical applications.
Keywords: Breast cancer detection, GLCM, Random Forest, Ductal carcinoma, Feature extraction, Machine learning, Symptom-based analysis, Classification.
|
DOI:
10.17148/IJARCCE.2025.14122
