Prediction of Lungs Cancer using Machine Learning Algorithms

Abstract

The primary cause of death is lung cancer, primarily due to the uncontrolled growth of malignant tumors in the lungs that can spread to the body’s other organs, posing serious health risks. Smoking is a major contributing factor. Early detection is crucial to prevent this deadly disease. In order to detect lungs cancer early on, we want to develop deep learning and machine learning algorithms. Such a model would help physicians make informed diagnostic decisions and determine the appropriate level of diagnostic intensity for patients. This method has the potential to significantly reduce treatment costs by enabling physicians to tailor treatment plans based on precise predictions, thereby avoiding unnecessary and expensive procedures. Our aim is to establish a sustainable model that forecasts lung cancer affectively. Our findings indicate that ResNet-50 surpasses other models, achieving an accuracy rate of 76%. In comparison Support Vector Machine (SVM), Logistic Regression (LR), and EfficientNet-B0 achieved accuracies of 73%, 64%, and 55%, respectively. This research underscores the potential of leveraging computer technology and machine learning methods to increase the precision of lung cancer diagnoses from CT scans