Missing data imputation in longitudinal cohort studies - Application of PLANN-ARD in breast cancer survival

Ana S. Fernandes; Ian H. Jarman; Terence A. Etchells; José M. Fonseca; Elia Biganzoli; Chris Bajdik; Paulo J G Lisboa

doi:10.1109/ICMLA.2008.106

Missing data imputation in longitudinal cohort studies - Application of PLANN-ARD in breast cancer survival

Ana S. Fernandes, Ian H. Jarman, Terence A. Etchells, José M. Fonseca, Elia Biganzoli, Chris Bajdik, Paulo J G Lisboa

Fondazione IRCCS Istituto Nazionale dei Tumori

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Missing values are common in medical datasets and may be amenable to data imputation when modelling a given data set or validating on an external cohort. This paper discusses model averaging over samples of the imputed distribution and extends this approach to generic non-linear modelling with the Partial Logistic Artificial Neural Network (PLANN) regularised within the evidence-based framework with Automatic Relevance Determination (ARD), The study then applies the imputation to external validation over new patient cohorts, considering also the case of predictions made for individual patients. A prognostic index is defined for the non-linear model and validation results show that 4 statistically significant risk groups identified at the 95% level of confidence from the modelling data, from Christie Hospital (n=931), retain good separation during external validation with data from the British Columbia Cancer Agency (n=4, 083).

Original language	English
Title of host publication	Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008
Pages	644-649
Number of pages	6
DOIs	https://doi.org/10.1109/ICMLA.2008.106
Publication status	Published - 2008
Event	7th International Conference on Machine Learning and Applications, ICMLA 2008 - San Diego, CA, United States Duration: Dec 11 2008 → Dec 13 2008

Other

Other	7th International Conference on Machine Learning and Applications, ICMLA 2008
Country/Territory	United States
City	San Diego, CA
Period	12/11/08 → 12/13/08

ASJC Scopus subject areas

Artificial Intelligence
Computer Science Applications
Software

Access to Document

10.1109/ICMLA.2008.106

Cite this

Fernandes, A. S., Jarman, I. H., Etchells, T. A., Fonseca, J. M., Biganzoli, E., Bajdik, C., & Lisboa, P. J. G. (2008). Missing data imputation in longitudinal cohort studies - Application of PLANN-ARD in breast cancer survival. In Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008 (pp. 644-649). [4725043] https://doi.org/10.1109/ICMLA.2008.106

Missing data imputation in longitudinal cohort studies - Application of PLANN-ARD in breast cancer survival. / Fernandes, Ana S.; Jarman, Ian H.; Etchells, Terence A. et al.

Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008. 2008. p. 644-649 4725043.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Fernandes, AS, Jarman, IH, Etchells, TA, Fonseca, JM, Biganzoli, E, Bajdik, C & Lisboa, PJG 2008, Missing data imputation in longitudinal cohort studies - Application of PLANN-ARD in breast cancer survival. in Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008., 4725043, pp. 644-649, 7th International Conference on Machine Learning and Applications, ICMLA 2008, San Diego, CA, United States, 12/11/08. https://doi.org/10.1109/ICMLA.2008.106

@inproceedings{9124e2c4840a46f2bbc4201733bfd723,

title = "Missing data imputation in longitudinal cohort studies - Application of PLANN-ARD in breast cancer survival",

abstract = "Missing values are common in medical datasets and may be amenable to data imputation when modelling a given data set or validating on an external cohort. This paper discusses model averaging over samples of the imputed distribution and extends this approach to generic non-linear modelling with the Partial Logistic Artificial Neural Network (PLANN) regularised within the evidence-based framework with Automatic Relevance Determination (ARD), The study then applies the imputation to external validation over new patient cohorts, considering also the case of predictions made for individual patients. A prognostic index is defined for the non-linear model and validation results show that 4 statistically significant risk groups identified at the 95% level of confidence from the modelling data, from Christie Hospital (n=931), retain good separation during external validation with data from the British Columbia Cancer Agency (n=4, 083).",

author = "Fernandes, {Ana S.} and Jarman, {Ian H.} and Etchells, {Terence A.} and Fonseca, {Jos{\'e} M.} and Elia Biganzoli and Chris Bajdik and Lisboa, {Paulo J G}",

year = "2008",

doi = "10.1109/ICMLA.2008.106",

language = "English",

isbn = "9780769534954",

pages = "644--649",

booktitle = "Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008",

note = "7th International Conference on Machine Learning and Applications, ICMLA 2008 ; Conference date: 11-12-2008 Through 13-12-2008",

}

TY - GEN

T1 - Missing data imputation in longitudinal cohort studies - Application of PLANN-ARD in breast cancer survival

AU - Fernandes, Ana S.

AU - Jarman, Ian H.

AU - Etchells, Terence A.

AU - Fonseca, José M.

AU - Biganzoli, Elia

AU - Bajdik, Chris

AU - Lisboa, Paulo J G

PY - 2008

Y1 - 2008

N2 - Missing values are common in medical datasets and may be amenable to data imputation when modelling a given data set or validating on an external cohort. This paper discusses model averaging over samples of the imputed distribution and extends this approach to generic non-linear modelling with the Partial Logistic Artificial Neural Network (PLANN) regularised within the evidence-based framework with Automatic Relevance Determination (ARD), The study then applies the imputation to external validation over new patient cohorts, considering also the case of predictions made for individual patients. A prognostic index is defined for the non-linear model and validation results show that 4 statistically significant risk groups identified at the 95% level of confidence from the modelling data, from Christie Hospital (n=931), retain good separation during external validation with data from the British Columbia Cancer Agency (n=4, 083).

AB - Missing values are common in medical datasets and may be amenable to data imputation when modelling a given data set or validating on an external cohort. This paper discusses model averaging over samples of the imputed distribution and extends this approach to generic non-linear modelling with the Partial Logistic Artificial Neural Network (PLANN) regularised within the evidence-based framework with Automatic Relevance Determination (ARD), The study then applies the imputation to external validation over new patient cohorts, considering also the case of predictions made for individual patients. A prognostic index is defined for the non-linear model and validation results show that 4 statistically significant risk groups identified at the 95% level of confidence from the modelling data, from Christie Hospital (n=931), retain good separation during external validation with data from the British Columbia Cancer Agency (n=4, 083).

UR - http://www.scopus.com/inward/record.url?scp=60649106740&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=60649106740&partnerID=8YFLogxK

U2 - 10.1109/ICMLA.2008.106

DO - 10.1109/ICMLA.2008.106

M3 - Conference contribution

AN - SCOPUS:60649106740

SN - 9780769534954

SP - 644

EP - 649

BT - Proceedings - 7th International Conference on Machine Learning and Applications, ICMLA 2008

T2 - 7th International Conference on Machine Learning and Applications, ICMLA 2008

Y2 - 11 December 2008 through 13 December 2008

ER -

Missing data imputation in longitudinal cohort studies - Application of PLANN-ARD in breast cancer survival

Abstract

Other

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this