A Robust Breast Cancer Classification Model using Extra-Trees Classifier for Histopathological Image

Abstract

Thousands of people die every quarter from breast cancer. Diagnosis and treatment at an early stage can drastically lower mortality rates. Traditional manual diagnosis, on the other hand, necessitates a large amount of labor by pathologists and is prone to diagnostic mistakes the longer they work. Rapid and accurate diagnosis are greatly aided by automatic histopathological image recognition. The biomedical industry has been drawn to Artificial Intelligence and its innovative methodologies because of its familiarity with the field's successes. Recent research has shown that AI can grasp details better than humans, leading to more accurate findings that aid professionals in making more informed judgments. This study presents the Extra-Tree classifier (ETC) for breast cancer image categorization. These findings demonstrate that ETC outperformed the other algorithms we examined for this data in terms of accuracy. Future researchers in the field of breast cancer will be able to use the findings of this study to guide their investigations and inform their efforts to boost the efficiency of certain algorithms.