Abstract
Globally, lung cancer is responsible for nearly one in five cancer deaths. The National Lung Screening Trial (NLST) demonstrated the efficacy of low-dose computed tomography (LDCT) to identify early-stage disease, setting the basis for widespread implementation of lung cancer screening programs. However, the specificity of LDCT lung cancer screening is suboptimal, with a significant false positive rate. Representing this imaging-based screening process as a sequential decision making problem, we combined multiple machine learning-based methods to learn a partially-observable Markov decision process that simultaneously optimizes lung cancer detection while enhancing test specificity. Using NLST data, we trained a dynamic Bayesian network as an observational model and used inverse reinforcement learning to discover a rewards function based on experts' decisions. Our resultant predictive model decreased the false positive rate while maintaining a high true positive rate at a level comparable to human experts. Our model also detected a number of lung cancers earlier.
Highlights
Lung cancer is the leading cause of cancer-related mortality, estimated to be responsible for 2.1 million deaths worldwide in 2018
In this work we present a novel approach that reduces the false positive (FP) rate associated with low-dose computed tomography (LDCT) lung cancer screening while maintaining a high true lung cancer detection rate
Our model reduces the FP rate in most screenings (t1, t2, and post-screening) compared to the experts while maintaining a high true positive (TP) rate for screening: at t0, true negative (TN) and TP rates are 2% lower and 3% lower than the physicians’; at t1, TN and TP rate are 1% higher and 3% higher; at t2, TN and TP rate are 4% higher and 4% lower; and in the post-screening period the partially-observable Markov decision process (POMDP)’s TN and TP rate are 3% higher and 8% higher than the experts’, respectively
Summary
Lung cancer is the leading cause of cancer-related mortality, estimated to be responsible for 2.1 million deaths worldwide in 2018. The findings of the National Lung Screening Trial (NLST) and more recent NELSON study [2], [3] support the implementation of lung cancer screening programs to identify individuals at high-risk for developing this disease, using low-dose computed tomography (LDCT) imaging to maximize initial detection [2], [4]. Of current concern is the disproportionately high false positive (FP) rate for LDCT screening: in the NLST, The associate editor coordinating the review of this article and approving it for publication was Vivek Kumar Sehgal. In this work we present a novel approach that reduces the FP rate associated with LDCT lung cancer screening while maintaining a high true lung cancer detection rate
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
