Classifying ovarian tumors using Bayesian multi-layer perceptrons and automatic relevance determination

A multi-center study

Ben Van Calster, Dirk Timmerman, Ian T. Nabney, Lil Valentin, Caroline Van Holsbeke, Sabine Van Huffel

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Ovarian masses are common and a good pre-surgical assessment of their nature is important for adequate treatment. Bayesian Multi-Layer Perceptrons (MLPs) using the evidence procedure were used to predict whether tumors are malignant or not. Automatic Relevance Determination (ARD) is used to select the most relevant of the 40+ available variables. Cross-validation is used to select an optimal combination of input set and number of hidden neurons. The data set consists of 1066 tumors collected at nine centers across Europe. Results indicate good performance of the models with AUC values of 0.93-0.94 on independent data. A comparison with a Bayesian perceptron model shows that the present problem is to a large extent linearly separable. The analyses further show that the number of hidden neurons specified in the ARD analyses for input selection may influence model performance.

Original languageEnglish
Title of host publicationAnnual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings
PublisherIEEE
Pages5342-5345
Number of pages4
ISBN (Print)1424400325, 9781424400324
DOIs
Publication statusPublished - 1 Dec 2006
Event28th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS'06 - New York, NY, United Kingdom
Duration: 30 Aug 20063 Sep 2006

Conference

Conference28th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS'06
CountryUnited Kingdom
CityNew York, NY
Period30/08/063/09/06

Fingerprint

Multilayer neural networks
Tumors
Neurons
Neural networks

Cite this

Van Calster, B., Timmerman, D., Nabney, I. T., Valentin, L., Van Holsbeke, C., & Van Huffel, S. (2006). Classifying ovarian tumors using Bayesian multi-layer perceptrons and automatic relevance determination: A multi-center study. In Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings (pp. 5342-5345). [4029269] IEEE. https://doi.org/10.1109/IEMBS.2006.260118
Van Calster, Ben ; Timmerman, Dirk ; Nabney, Ian T. ; Valentin, Lil ; Van Holsbeke, Caroline ; Van Huffel, Sabine. / Classifying ovarian tumors using Bayesian multi-layer perceptrons and automatic relevance determination : A multi-center study. Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings. IEEE, 2006. pp. 5342-5345
@inproceedings{915ade3fd0474b1ea5e42feaf816dcea,
title = "Classifying ovarian tumors using Bayesian multi-layer perceptrons and automatic relevance determination: A multi-center study",
abstract = "Ovarian masses are common and a good pre-surgical assessment of their nature is important for adequate treatment. Bayesian Multi-Layer Perceptrons (MLPs) using the evidence procedure were used to predict whether tumors are malignant or not. Automatic Relevance Determination (ARD) is used to select the most relevant of the 40+ available variables. Cross-validation is used to select an optimal combination of input set and number of hidden neurons. The data set consists of 1066 tumors collected at nine centers across Europe. Results indicate good performance of the models with AUC values of 0.93-0.94 on independent data. A comparison with a Bayesian perceptron model shows that the present problem is to a large extent linearly separable. The analyses further show that the number of hidden neurons specified in the ARD analyses for input selection may influence model performance.",
author = "{Van Calster}, Ben and Dirk Timmerman and Nabney, {Ian T.} and Lil Valentin and {Van Holsbeke}, Caroline and {Van Huffel}, Sabine",
year = "2006",
month = "12",
day = "1",
doi = "10.1109/IEMBS.2006.260118",
language = "English",
isbn = "1424400325",
pages = "5342--5345",
booktitle = "Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings",
publisher = "IEEE",
address = "United States",

}

Van Calster, B, Timmerman, D, Nabney, IT, Valentin, L, Van Holsbeke, C & Van Huffel, S 2006, Classifying ovarian tumors using Bayesian multi-layer perceptrons and automatic relevance determination: A multi-center study. in Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings., 4029269, IEEE, pp. 5342-5345, 28th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS'06, New York, NY, United Kingdom, 30/08/06. https://doi.org/10.1109/IEMBS.2006.260118

Classifying ovarian tumors using Bayesian multi-layer perceptrons and automatic relevance determination : A multi-center study. / Van Calster, Ben; Timmerman, Dirk; Nabney, Ian T.; Valentin, Lil; Van Holsbeke, Caroline; Van Huffel, Sabine.

Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings. IEEE, 2006. p. 5342-5345 4029269.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Classifying ovarian tumors using Bayesian multi-layer perceptrons and automatic relevance determination

T2 - A multi-center study

AU - Van Calster, Ben

AU - Timmerman, Dirk

AU - Nabney, Ian T.

AU - Valentin, Lil

AU - Van Holsbeke, Caroline

AU - Van Huffel, Sabine

PY - 2006/12/1

Y1 - 2006/12/1

N2 - Ovarian masses are common and a good pre-surgical assessment of their nature is important for adequate treatment. Bayesian Multi-Layer Perceptrons (MLPs) using the evidence procedure were used to predict whether tumors are malignant or not. Automatic Relevance Determination (ARD) is used to select the most relevant of the 40+ available variables. Cross-validation is used to select an optimal combination of input set and number of hidden neurons. The data set consists of 1066 tumors collected at nine centers across Europe. Results indicate good performance of the models with AUC values of 0.93-0.94 on independent data. A comparison with a Bayesian perceptron model shows that the present problem is to a large extent linearly separable. The analyses further show that the number of hidden neurons specified in the ARD analyses for input selection may influence model performance.

AB - Ovarian masses are common and a good pre-surgical assessment of their nature is important for adequate treatment. Bayesian Multi-Layer Perceptrons (MLPs) using the evidence procedure were used to predict whether tumors are malignant or not. Automatic Relevance Determination (ARD) is used to select the most relevant of the 40+ available variables. Cross-validation is used to select an optimal combination of input set and number of hidden neurons. The data set consists of 1066 tumors collected at nine centers across Europe. Results indicate good performance of the models with AUC values of 0.93-0.94 on independent data. A comparison with a Bayesian perceptron model shows that the present problem is to a large extent linearly separable. The analyses further show that the number of hidden neurons specified in the ARD analyses for input selection may influence model performance.

UR - http://www.scopus.com/inward/record.url?scp=34047133238&partnerID=8YFLogxK

UR - https://ieeexplore.ieee.org/document/4463010

U2 - 10.1109/IEMBS.2006.260118

DO - 10.1109/IEMBS.2006.260118

M3 - Conference contribution

SN - 1424400325

SN - 9781424400324

SP - 5342

EP - 5345

BT - Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings

PB - IEEE

ER -

Van Calster B, Timmerman D, Nabney IT, Valentin L, Van Holsbeke C, Van Huffel S. Classifying ovarian tumors using Bayesian multi-layer perceptrons and automatic relevance determination: A multi-center study. In Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings. IEEE. 2006. p. 5342-5345. 4029269 https://doi.org/10.1109/IEMBS.2006.260118