A Combınatıon of BI-Clusterıng and Hybrıd Boostıng Algorıthm for Breast Tumor Classıfıcatıon

Abstract

Breast malignancy is generally accepted as one of the most important causes of female mortality worldwide. A new study sponsored by PC that works with a human-on-top approach is proposed to enable clinicians and doctors to differentiate malignant and benign tumors of the bosom using gradient-boosting ultrasound computations. In order to achieve the performance a performance rating scheme focused on the Breast Imaging Monitoring and Measurement Framework (BI-RADS) lexicon was used. Since other methods used to classify breast tumors are similar, if there are any unusual cases or new cases to be identified using  the proposed  Hybrid  boosting  algorithm, it will be difficult to pick the right classifier. The work is conducted to explore and incorporate suitable variations of tumor characteristics into a robust classifier, thereby opening the way for greater precision. The suggested approach has been tested using an ultrasonic dataset of breast tumor occurrences and gives a good result   and also an ideal way to support enormous promise in clinical applications