The wisdom of the commons

Ensemble tree classifiers for prostate cancer prognosis

James A. Koziol, Anne C. Feng, Zhenyu Jia, Yipeng Wang, Steven Goodison, Michael Mcclelland, Dan Mercola

Research output: Contribution to journalArticle

20 Citations (Scopus)

Abstract

Motivation: Classification and regression trees have long been used for cancer diagnosis and prognosis. Nevertheless, instability and variable selection bias, as well as overfitting, are well-known problems of tree-based methods. In this article, we investigate whether ensemble tree classifiers can ameliorate these difficulties, using data from two recent studies of radical prostatectomy in prostate cancer. Results: Using time to progression following prostatectomy as the relevant clinical endpoint, we found that ensemble tree classifiers robustly and reproducibly identified three subgroups of patients in the two clinical datasets: non-progressors, early progressors and late progressors. Moreover, the consensus classifications were independent predictors of time to progression compared to known clinical prognostic factors.

Original languageEnglish (US)
Pages (from-to)54-60
Number of pages7
JournalBioinformatics
Volume25
Issue number1
DOIs
StatePublished - Jan 1 2009
Externally publishedYes

Fingerprint

Prostate Cancer
Prognosis
Prostatectomy
Prostatic Neoplasms
Ensemble
Classifiers
Classifier
Progression
Selection Bias
Classification and Regression Trees
Prognostic Factors
Overfitting
Variable Selection
Predictors
Cancer
Subgroup
Neoplasms
Datasets

ASJC Scopus subject areas

  • Statistics and Probability
  • Biochemistry
  • Molecular Biology
  • Computer Science Applications
  • Computational Theory and Mathematics
  • Computational Mathematics

Cite this

The wisdom of the commons : Ensemble tree classifiers for prostate cancer prognosis. / Koziol, James A.; Feng, Anne C.; Jia, Zhenyu; Wang, Yipeng; Goodison, Steven; Mcclelland, Michael; Mercola, Dan.

In: Bioinformatics, Vol. 25, No. 1, 01.01.2009, p. 54-60.

Research output: Contribution to journalArticle

Koziol, JA, Feng, AC, Jia, Z, Wang, Y, Goodison, S, Mcclelland, M & Mercola, D 2009, 'The wisdom of the commons: Ensemble tree classifiers for prostate cancer prognosis', Bioinformatics, vol. 25, no. 1, pp. 54-60. https://doi.org/10.1093/bioinformatics/btn354
Koziol, James A. ; Feng, Anne C. ; Jia, Zhenyu ; Wang, Yipeng ; Goodison, Steven ; Mcclelland, Michael ; Mercola, Dan. / The wisdom of the commons : Ensemble tree classifiers for prostate cancer prognosis. In: Bioinformatics. 2009 ; Vol. 25, No. 1. pp. 54-60.
@article{ee94b87ddd0a4fb6abef651c2251c7f3,
title = "The wisdom of the commons: Ensemble tree classifiers for prostate cancer prognosis",
abstract = "Motivation: Classification and regression trees have long been used for cancer diagnosis and prognosis. Nevertheless, instability and variable selection bias, as well as overfitting, are well-known problems of tree-based methods. In this article, we investigate whether ensemble tree classifiers can ameliorate these difficulties, using data from two recent studies of radical prostatectomy in prostate cancer. Results: Using time to progression following prostatectomy as the relevant clinical endpoint, we found that ensemble tree classifiers robustly and reproducibly identified three subgroups of patients in the two clinical datasets: non-progressors, early progressors and late progressors. Moreover, the consensus classifications were independent predictors of time to progression compared to known clinical prognostic factors.",
author = "Koziol, {James A.} and Feng, {Anne C.} and Zhenyu Jia and Yipeng Wang and Steven Goodison and Michael Mcclelland and Dan Mercola",
year = "2009",
month = "1",
day = "1",
doi = "10.1093/bioinformatics/btn354",
language = "English (US)",
volume = "25",
pages = "54--60",
journal = "Bioinformatics",
issn = "1367-4803",
publisher = "Oxford University Press",
number = "1",

}

TY - JOUR

T1 - The wisdom of the commons

T2 - Ensemble tree classifiers for prostate cancer prognosis

AU - Koziol, James A.

AU - Feng, Anne C.

AU - Jia, Zhenyu

AU - Wang, Yipeng

AU - Goodison, Steven

AU - Mcclelland, Michael

AU - Mercola, Dan

PY - 2009/1/1

Y1 - 2009/1/1

N2 - Motivation: Classification and regression trees have long been used for cancer diagnosis and prognosis. Nevertheless, instability and variable selection bias, as well as overfitting, are well-known problems of tree-based methods. In this article, we investigate whether ensemble tree classifiers can ameliorate these difficulties, using data from two recent studies of radical prostatectomy in prostate cancer. Results: Using time to progression following prostatectomy as the relevant clinical endpoint, we found that ensemble tree classifiers robustly and reproducibly identified three subgroups of patients in the two clinical datasets: non-progressors, early progressors and late progressors. Moreover, the consensus classifications were independent predictors of time to progression compared to known clinical prognostic factors.

AB - Motivation: Classification and regression trees have long been used for cancer diagnosis and prognosis. Nevertheless, instability and variable selection bias, as well as overfitting, are well-known problems of tree-based methods. In this article, we investigate whether ensemble tree classifiers can ameliorate these difficulties, using data from two recent studies of radical prostatectomy in prostate cancer. Results: Using time to progression following prostatectomy as the relevant clinical endpoint, we found that ensemble tree classifiers robustly and reproducibly identified three subgroups of patients in the two clinical datasets: non-progressors, early progressors and late progressors. Moreover, the consensus classifications were independent predictors of time to progression compared to known clinical prognostic factors.

UR - http://www.scopus.com/inward/record.url?scp=58049211939&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=58049211939&partnerID=8YFLogxK

U2 - 10.1093/bioinformatics/btn354

DO - 10.1093/bioinformatics/btn354

M3 - Article

VL - 25

SP - 54

EP - 60

JO - Bioinformatics

JF - Bioinformatics

SN - 1367-4803

IS - 1

ER -