Breast Tumor Detection and Classification Using ABC-ELM Algorithm

Abstract

Analyzing medical images and deciding on health issues require a hybrid intelligence system. Today, cancer is the illness that kills the most people worldwide. The most common type of tumor sickness is breast cancer. The previously employed algorithms for detecting and classifying breast tumors kept getting better until they produced successful results. Machine learning techniques for detecting and classifying breast tumors have been the subject of numerous studies. For breast tumor identification and classification, we proposed the Artificial Bee Colony Extreme Learning Machine (ABC-ELM) algorithm. We aim to improve the breast tumor identification and classification algorithm using the Morlet wavelet transform for feature extraction and tumor segmentation. Breast images are classified as malignant or non-cancerous by extracting pertinent features and improving the ELM classifier's parameters using the ABC algorithm. As a result, the proposed method performs best compared to earlier comparative research, achieving an accuracy of 98.2%, Sensitivity is 98.5%, and precision 98.5%.