Leveraging Machine Learning and Deep Learning to Enhance Lean Operations in Healthcare: A Focus on Lung Cancer Detection




De La Rosa, Kevin

Journal Title

Journal ISSN

Volume Title



As cancer ranks within the top five causes of death within the United States, the current cancer environment, applications of lean methodologies in the healthcare industry, and the implementation of artificial intelligence and machine learning to support cancer treatment, patient's experiences, and oncology operations is explored. Statistical analysis is then performed on a lung cancer patient dataset to understand the correlation the variables have to cancer diagnosis. Various artificial intelligence models such as Random Forest (RF), Convolutional Neural Networks (CNN), Logistic Regression (LR), Extreme Gradient Boosting (XGBoost), and Multilayer Perceptron Neural Network (MLP-NN) are then applied to the dataset to evaluate model accuracy and identify if the application can improve oncology centers operational and treatment efficiency. XGBoost with and without Principal Component Analysis (PCA), Logistic Regression, Random Forest, and MLP-NN with and without PCA achieved an accuracy of 100%, with LR with PCA (98.93%), and CNN (96.27%) following. These high accuracies confirm the implementation of artificial intelligence within the healthcare organization can be successful in supporting diagnosis predictions and enhancing lean operations.



Artificial Intelligence, Computational Pathology, Healthcare, Lean Six Sigma, Lung Cancer, Oncology



Mechanical Engineering