.

ISSN 2063-5346
For urgent queries please contact : +918130348310

Evaluation of SVM and NN Machine Learning Classifiers in Thyroid Tumor Detection system

Main Article Content

Dr . M. Dharani, G. Sreehitha, Dr. Maganti Venkatesh, Dr. Dumpa Prasad, Mr. Nayani Sateesh, Dr. Umamaheswararao Batta
» doi: 10.31838/ecb/2023.12.sa1.077

Abstract

Cancer is one of the world's most hazardous diseases, and it is particularly effective in women. As a result, our primary goal must be to cure cancer through scientific inquiry, with early identification of cancer as a secondary goal. Diagnosis can aid in the complete removal of cancer. Cancer is critical for improving survival rates. As a result, a reliable diagnosis and detection procedure is required. Medical practitioners will benefit greatly from automatic detection techniques. There are several approaches for cancer detection that have been proposed. However, this is not a simple task due to various uncertainties in mammography detection. Mammograms are images created by a radiologist using a machine. The doctor examines these mammograms and diagnoses the cancer for future therapy. Because all general hospitals lack specialists, patients have had to wait for their results. As a result, waiting for a breast cancer diagnosis may take some time. This delay may have caused the cancer to spread, lowering the patient's chances of survival. Machine Learning (ML) methods can be utilized to produce tools for doctors which can be used as an efficient system for early cancer detection and diagnosis, considerably improving patient survival rates. This does not imply that a computer can replace an expert or a physician, but rather that a computer can help an expert better grasp a case and produce findings more quickly. K-Nearest Neighbors, Support Vector Machine (SVM), and Neural Network (NN) are examples of machine learning algorithms that help us tackle this challenge by producing solutions with great sensitivity and accuracy. Various metrics such as specificity, sensitivity, and accuracy are used to evaluate cancer images in this study.

Article Details