COMPARATIVE ANALYSIS OF LUNG CANCER DETECTION USING MACHINE LEARNING AND BIOSENSORS

Abstract

Each cell in the human body obtains adequate oxygen thanks to the lungs, which act as the breathing's control centre. To prevent contaminants and diseases from reaching the body filters the air simultaneously. Diseases inflammatory conditions, or more severe issues like the emergence of an aggressive tumor can have an effect on the lungs.  The prediction of lung cancer was examined using a variety of classification algorithms, including SVM, Decision trees, Naive Bayes, and Logistic Regression. This study's main objective is to determine how well classification algorithms work for detecting lung cancer early on. This study's main goal is to compare the degrees of accuracy across multiple artificial intelligence algorithms. The many models used by researchers were given out, and they have a few restrictions and their disadvantages were set out, in order to determine the accuracy levels of different classifiers. A thorough review of the literature revealed that some classifiers have poor accuracy while others have greater accuracy but haven't quite achieved 100%. The several biosensors for finding lung cancer biomarkers are described in this review. The several documented indicators for lung cancer are first briefly described. The advancements in developing sensor technology for specific, secure, and specific identification of lung cancer indicators are then carefully detailed, with an emphasis on the key clinical biomarkers such as MicroRNA, CA 125, Tumor-Associated Antigens (TAAs), and SCCA. The potential for developing effective sensor technology for the prompt identification of cancer in the lungs is then discussed, along with the existing challenges. Tumor markers are biochemical characteristics that can reflect the existence and evolution of cancer. They exhibit sensitivity, convenience, and affordability in the development of biosensors, making them excellent candidates for the creation of lung cancer detection biosensors.