International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 09 Issue: 12 | Dec 2022
p-ISSN: 2395-0072
www.irjet.net
Logistic Regression Model for Predicting the Malignancy of Breast Cancer Muneeba Ahmed1 1Systems
Engineer, Infosys Limited, India ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - In today's modern environment, recognising
Numerous methods for estimating breast cancer have been identified in the last year. During biopsy screening, the breast tissues are used for the biopsies. Although the testing yields more trustworthy results, the method for collecting breast biopsies is incredibly painful and pitiable [2]. The majority of patients are not interested in this testing as a result. Since mammography produces 2D projection images of the breast, it is the most widely used method for estimating breast cancer. The two most frequently utilised mammogram techniques are digital mammography and screen-film mammography [3]. Screenfilm mammography is used on female breasts that are asymptomatic. It takes roughly 20 minutes to do a traditional mammogram. Benign cancer cannot be found with this method. Digital mammography offers a solution to the screening mammography problem. It is connected to a computing equipment since a computer is where digital mammography data is saved. Digital mammography uses image processing techniques to enhance the quality of the images that are recorded. Digital mammography performs better for incorrectly diagnosed samples. Magnetic resonance imaging, another common technique, is primarily used to find breast cancer [4]. The MRI is a challenging procedure. Additionally, certain malignancies that mammography would have detected could be missed. In women who have been given a breast cancer diagnosis, MRI is used to measure the breast's actual size and spot numerous disorders in the breast.
breast cancer is critical. Breast cancer is one of the most serious tumours that can affect women, and it can be fatal. Breast cancer is classified into two types: benign (noncancerous) and malignant (cancerous). Machine learning is the process through which a machine learns increasingly on its own. The ML model is a mathematical technique used in artificial intelligence. A computer that thinks for itself and mimics human intelligence is referred to as artificial intelligence. Just like a human, the computer improves at its work as it gets "experience." There are several Machine Learning approaches available for analysing breast cancer data. This paper describes a Machine Learning model for diagnosing breast cancer. Logistic Regression model is used for detecting breast cancer. This algorithm falls under the category of supervised machine learning. Key Words: Breast Cancer, Artificial Intelligence, Machine Learning, Logistic Regression
1. INTRODUCTION Breast cancer refers to the uncontrolled cell development in the breast. Both men and women can get breast cancer, but women are more likely to have it. Breast cancer has been one of the main causes of female mortality when compared to other malignancies. Breast cancer symptoms include changes in the breast's size and form, the thickness of the tissue around the breast, as well as crust, scales, and redness of the skin. Changes in environmental variables, hormones, and lifestyle lead to breast cancer, which raises the risk factor. The lymphatic vessels allow lymphatic fluid from the breast to pass through. If the breast contains cancerous cells, they go into the lymphatic vessels and start to multiply in the lymph nodes. Although many breast cancer patients have no symptoms at all, breast cancer is typically discovered after the beginning of symptoms. To prevent mortality, early detection of breast cancer is crucial. For the ability to detect breast cancer in its early stages, earlier therapy is required. A reliable and efficient diagnostic method that enables clinicians to differentiate between benign and malignant breast tumours is required for early identification. For the current medical issue, the automated identification of breast cancer is significant. It is crucial to create an efficient and reliable diagnostic strategy. Clinical applications face a major problem with clinical diagnosis. Breast cancer data classifications can be used to predict the outcomes of specific diseases and to determine the genetic activity of tumours [1].
© 2022, IRJET
|
Impact Factor value: 7.529
In the past year, machine learning techniques have been used more and more in prediction, especially in the field of medicine [5]. It gives systems the ability to learn from the past in order to extrapolate intricate insights from massive data sets. In a variety of clinical settings, these methods are most frequently employed to identify and classify malignancies. In order to diagnose and cure breast cancer, machine learning has been used first and foremost [6].
2. RELATED WORK This section discusses some of the related research on machine learning-based breast cancer diagnosis that has been conducted in the past. S.Vasundhara, B.V. Kiranmayee, and Chalumuru Suresh [7] proposed employing several machine learning methods to classify mammography pictures as benign, malignant, or normal. A comparison of Support Vector Machines, Convolutional Neural Networks, and Random Forest is
|
ISO 9001:2008 Certified Journal
|
Page 440