Comparison study on Lung Cancer Risk Assignment using Machine Learning and Deep Learning

Abstract

Lung tumour is the expansion of malignant cells in the lungs. The rising frequency of cancer has brought about a rise in the rate of death for both women and men. Uncontrolled cell multiplication in the lungs is a condition known as lung tumor. Although it’s not possible to prevent lung cancer, but risk can be decreased. To keep lung issues from developing into long-term, serious illnesses, early diagnosis and treatment are essential. SVM, Logistic regression, KNN, and Random Forest Decision trees were among the techniques for classification used to assess the lung cancer risk prediction. further enhanced the technique's accuracy by using boosting classifiers like gradient, Ada, and XG classifiers. By assessing the effectiveness of classification algorithms, the primary goal of this study is to diagnose lung cancer early enough. In this research, we present an analysis comparing Machine learning methods with Deep Learning-based Computer Assisted Diagnosis methods. To tackle these issues, our study shows good results in lung disease identification even with a smaller dataset, offering a workable solution to the problems in the area of medical image processing. When compared to other machine learning algorithms, SVM has produced more precise results. Furthermore, we provide an outline of the benefits and drawbacks of the current set of lung cancer identification algorithms.