Survival analysis in cancer using a partial logistic neural network model with Bayesian regularisation framework: a validation study
Source: International Journal of Knowledge Engineering and Soft Data Paradigms, Volume 1, Number 3, 3 October 2009 , pp. 277-295(19)
Publisher: Inderscience Publishers
Abstract:This paper describes a multicentre longitudinal cohort study to evaluate the predictive accuracy of a regularised Bayesian neural network model in a prognostic application. The study sample (n = 5442) comprises subjects treated with intraocular melanoma in two different centres in Liverpool and Paris. External validation was carried out by fitting the model to the data from Liverpool set and predicting for the data from Paris. The performance of the model in out‐of‐sample prediction was assessed statistically for discrimination of outcomes and calibration. It was also evaluated clinically by comparing against the accepted TNM staging system. The model had good discrimination with Harrell's C index > 0.7 up to ten years of follow‐up. Calibration results were also good up to ten years using a Hosmer‐Lemeshow type analysis (p > 0.05). The paper: 1) deals with the issue of missing data using methods that are well accepted in the literature; 2) proposes a framework for externally validating machine learning models applied to survival analysis; 3) applies accepted methods for dealing with missing data; 4) proposes an alternative staging system based on the model. The new staging system, which takes into account histopathologic information, has several advantages over the existing staging system.
Keywords: COMPUTING AND MATHEMATICS JOURNALS; Computing Science, Applications and Software; EDUCATION, KNOWLEDGE AND LEARNING JOURNALS; Electronic Systems, Control and Artificial Intelligence; Knowledge Studies; TECHNICAL JOURNALS
Document Type: Research Article
Affiliations: 1: Department Medical Physics and Clinical Engineering, Royal Liverpool University Hospital, Daulby Street, Liverpool L7 8XP, UK. 2: Department Electrical Engineering and Electronics, University of Liverpool, Liverpool L69 3BX, UK. 3: School of Mathematics and Computing Sciences, Liverpool John Moores University, Byrom Street, Liverpool L3 3AF, UK. 4: Department Ophthalmology, Institut Curie, 26 Rue d 5: ulm, Paris 75005, France.
Publication date: October 3, 2009
- International Journal of Knowledge Engineering and Soft Data Paradigms reports the most recent research results in the areas of knowledge engineering and soft data analysis. Knowledge engineering, in contrast to traditional engineering techniques, involves mimicking the performance of human experts in a limited sense. Knowledge engineering with soft data paradigms offers very significant attributes such as learning, autonomy and self-organisation.