Prediction of lung cancer incidence on the low-dose computed tomography arm of the National Lung Screening Trial: A dynamic Bayesian network

Highlights • Dynamic Bayesian Network (DBN) built using a lung cancer staging state-space model. • Use of resampling techniques to address data imbalance. • Results are comparable to experts’ decisions. • Similar structure and performance between learned and expert-derived DBNs.

Saved in:
Bibliographic Details
Published in:Artificial intelligence in medicine 2016-09, Vol.72, p.42-55
Main Authors: Petousis, Panayiotis, Han, Simon X, Aberle, Denise, Bui, Alex A.T
Format: Article
Language:English
Subjects:
Annual NLST cancer risk
Bayes Theorem
Cancer incidence
Clinical Trials as Topic
Dynamic Bayesian networks
Early Detection of Cancer
Expert-driven networks
Humans
Incidence
Individualized lung cancer screening
Internal Medicine
Lung Neoplasms - diagnosis
Lung Neoplasms - epidemiology
Lung stage cancer state-space
Mass Screening
Models, Theoretical
Neoplasm Staging
Other
ROC Curve
Smoking
Structure learning
Tomography, X-Ray Computed
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Highlights • Dynamic Bayesian Network (DBN) built using a lung cancer staging state-space model. • Use of resampling techniques to address data imbalance. • Results are comparable to experts’ decisions. • Similar structure and performance between learned and expert-derived DBNs.
ISSN:0933-3657
1873-2860
DOI:10.1016/j.artmed.2016.07.001