Determine the Degree of Malignancy in Breast Cancer using Machine Learning
Authors:
- 11 October 2023
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 483-487
Abstract
It is predicted that 2.1 million women were given a diagnosis of breast cancer for the first time in 2018, and that 626,679 women who had breast cancer passed away because of the disease. Breast cancer is the most common form of cancer among women globally. Even though there is a wide variety of approaches to determining the presence or absence of malignancy, there are several procedures that raise concerns due to their varying degrees of accuracy in diagnosis [3]. As a result of the precision and speed of computer-assisted diagnostics, machine learning is receiving an increasing amount of attention for its potential role in aiding with illness prediction in this field. For this work, we used several different machine learning methods to investigate breast cancer malignancies. To evaluate whether breast tissue cells acquired by fine needle aspiration (FNA) were malignant, these methods were used to a diagnostic dataset of characteristics from those cells. We observed that the k-nearest neighbor (KNN) model had the greatest performance of all the models that were evaluated and scored the maximum accuracy. This was determined by the creation of confusion matrices and the evaluation of several different metrics (97.66 percent). We were also successful in developing a simpler generalized additive model (GAM) that depended on just three characteristics but achieved an accuracy level that was like the original model (95.91 percent). R Studio was used for all the tests that were conducted.Keywords
This publication has 10 references indexed in Scilit:
- A Systematic Review of Real-time Fine-needle Aspiration Biopsy Methods for Soft TissuesIETE Technical Review, 2021
- Breast cancerNature Reviews Disease Primers, 2019
- Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countriesCA: A Cancer Journal for Clinicians, 2018
- The Screening, Diagnosis, Treatment, and Follow-Up of Breast CancerDeutsches Ärzteblatt international, 2018
- Detection and Classification of Cancer from Microscopic Biopsy Images Using Clinically Significant and Biologically Interpretable FeaturesJournal of Medical Engineering, 2015
- Cancer Incidence in Five Continents: Inclusion criteria, highlights from Volume X and the global status of cancer registrationInternational Journal of Cancer, 2015
- Effect of Three Decades of Screening Mammography on Breast-Cancer IncidenceNew England Journal of Medicine, 2012
- The American Cancer Society and the American Health Care SystemThe Oncologist, 2011
- Breast cancer detection from FNA using SVM with different parameter tuning systems and SOM–RBF classifierJournal of the Franklin Institute, 2007
- Nuclear feature extraction for breast tumor diagnosis1993