A Novel Bayesian Framework For Multi-State Disease Progression Of Lung Cancer

Abstract

CT screening has been commonly used to identify and diagnose lung cancer in its early stages. CT has been shown in clinical studies to reduce lung cancer mortality by 20% as compared to plain chest radiography; however, existing CT screening services face obstacles such as high over diagnosis rates, high costs, and elevated radiation exposure.The study develops computer and deep learning models for predictive lung cancer diagnosis and disease progression prediction in an effort to solve these difficulties. Using a symmetric chain code method and a machine learning system, a novel lung segmentation approach was first developed. The lung nodules connected to the lung wall are included in this process, which minimises over-segmentation error. Finally, to predict the inter disease progression of lung cancer, a Bayesian method was coupled with a prolonged Markov model.The resultant model calculates specific lung cancer state transition data, which can be used to make customised screening recommendations. Extensive trials and results have shown the efficacy of these approaches, paving the way for current CT screening systems to be optimised and improved.

Design Computer-Aided Diagnosis System Based on Chest CT Evaluation of Pulmonary Nodules

“…At present, most of the researches on enhancement algorithms for suspected nodules mostly use structural shape enhancement filtering methods. The specific principle is to use the second-order partial differential information of the image to establish the Hessian matrix and use different eigenvalues to have a strong response to different scales of spherical and linear structures and to construct filters to select nodules and blood vessels, respectively [ 16 , 17 ]. However, this method has limitations: (1) the X-shaped or Y-shaped intersection of blood vessels cannot be effectively identified; (2) the target shape and structure selected by filtering are not completely accurate, especially for burrs, nodules; (3) since the Hessian matrix contains high-order partial differential information, the selected target may have false edges caused by noise points.…”