Using Bayesian neural networks with ARD input selection to detect malignant ovarian masses prior to surgery

Ben Van Calster, Dirk Timmerman, Ian T. Nabney, Lil Valentin, Antonia C. Testa, Caroline Van Holsbeke, Ignace Vergote, Sabine Van Huffel

Research output: Contribution to journalArticle (Academic Journal)peer-review

10 Citations (Scopus)

Abstract

In this paper, we applied Bayesian multi-layer perceptrons (MLP) using the evidence procedure to predict malignancy of ovarian masses in a large (n = 1,066) multi-centre data set. Automatic relevance determination (ARD) was used to select the most relevant inputs. Fivefold cross-validation (5CV) and repeated 5CV was used to select the optimal combination of input set and number of hidden neurons. Results indicate good performance of the models with area under the receiver operating characteristic curve values of 0.93-0.94 on independent data. Comparison with a linear benchmark model and a previously developed logistic regression model shows that the present problem is very well linearly separable. A resampling analysis further shows that the number of hidden neurons specified in the ARD analyses for input selection may influence model performance. This paper shows that Bayesian MLPs, although not frequently used, are a useful tool for detecting malignant ovarian tumours.
Original languageEnglish
Pages (from-to)489-500
Number of pages12
JournalNeural Computing and Applications
Volume17
Issue number5-6
DOIs
Publication statusPublished - 1 Oct 2008

Keywords

  • Automatic relevance determination, Bayesian evidence framework, Multi-layer perceptrons, Netlab, Ovarian tumour classification, Ultrasound

Fingerprint Dive into the research topics of 'Using Bayesian neural networks with ARD input selection to detect malignant ovarian masses prior to surgery'. Together they form a unique fingerprint.

Cite this