Machine Learning Applications in Healthcare: A Deep Dive into CNN and ML Models for Breast Cancer, Lung Cancer, and COVID-19 Diagnosis

Abstract

Machine learning (ML) models have proven to be important tools in the diagnosis of multiple diseases in medicine. In this study, various algorithms like logistic regression, support vector machine (SVM), decision trees, typical models like XGboost or AdaBoost, and combination of neural networks for clustering is used.VGG-16, VGG etal., Deep Learning Architecture – 19, ResNet-50, and ResNet-101. Before being used for diagnostic, medical images are preprocessed to improve quality and achieve better outcome. Because CNNs have a deep hierarchical structure, each of its successful models adds a special contribution to the classification process of CNNs and the state-of-the-art inference variants. This study attempts to contrast the results and result performance of these models for breast cancer diagnosis, lung cancer diagnosis and COVID-19 diagnosis without specifying any accuracy. Machine learning models’ integration into health analytics includes disease detection and treatment. Diagnostic criteria are different between each of the models that are used. Logistic regression, SVM, Decision Trees, XGBoost, and AdaBoost are a traditional model that helps us classify what we are doing and provide an efficient and effective result of production. Deep learning models, such as CNN, VGG-16, VGG-19, are on the other hand. ResNet-50 and ResNet-101 are able to do well processing high quality objects like medical images because they have features. It has powerful functionality and deep architecture. Medical applications are unique and capture the complex patterns in images. Assess cost and confirm performance.