Paper Title
Breast Cancer Classification and Segmentation using Boosting Algorithms and U-Net
Abstract
The most common disease among women globally is breast cancer, which is treatable in 70-80 percent of
individuals with early-stage cases. Various machine learning approaches are used for finding the cancer. But compared to
machine learning techniques, the ensemble classifier provides a good results. Because to its benefits in modelling a vital
feature detection from complicated BC datasets, machine learning (ML) is frequently utilised in the categorization of breast
cancer (BC) patterns. In this research, we propose an ensemble of classifiers-based system for the automatic detection of
breast cancer. We begin by reviewing various machine learning (ML) algorithms and ML algorithm ensembles. For the
purpose of automatic Breast cancer detection, we offer an overview of Machine learning techniques, and ensemble of several
classifiers. Additionally, we present and contrast multiple ensemble and other tested ML-based model variations. When
noise, contrast adjustment, and distortions are present, segmentation becomes tedious. Before to segmentation, preprocessing
is done to improve contrast and remove extraneous data from the image. The U-Net model will enhance
computing when there are powerful GPUs present, facilitating the training of networks with additional layers. However,
recent studies have shown that early use of preprocessing techniques would undoubtedly increase accuracy.
Keywords - Breast Cancer, Machine Learning, Ensemble Learning, Segmentation, U-Net, mask
Author - T. Anil Raju, S. Gayathri, V. Jayanth, T.Pawan Kalyan
Published : Volume-10,Issue-5 ( May, 2023 )
DOIONLINE Number - IJAECS-IRAJ-DOIONLINE-19743
View Here
|
|
| |
|
PDF |
| |
Viewed - 14 |
| |
Published on 2023-08-24 |
|