A critical appraisal of logistic regression-based nomograms, artificial neural networks, classification and regression-tree models, look-up tables and risk-group stratification models for prostate cancer

Felix K H Chun, Pierre I. Karakiewicz, Alberto Briganti, Jochen Walz, Michael W. Kattan, Hartwig Huland, Markus Graefen

Research output: Contribution to journalArticle

Abstract

OBJECTIVE: To evaluate several methods of predicting prostate cancer-related outcomes, i.e. nomograms, look-up tables, artificial neural networks (ANN), classification and regression tree (CART) analyses and risk-group stratification (RGS) models, all of which represent valid alternatives. METHODS: We present four direct comparisons, where a nomogram was compared to either an ANN, a look-up table, a CART model or a RGS model. In all comparisons we assessed the predictive accuracy and performance characteristics of both models. RESULTS: Nomograms have several advantages over ANN, look-up tables, CART and RGS models, the most fundamental being a higher predictive accuracy and better performance characteristics. CONCLUSION: These results suggest that nomograms are more accurate and have better performance characteristics than their alternatives. However, ANN, look-up tables, CART analyses and RGS models all rely on methodologically sound and valid alternatives, which should not be abandoned.

Original languageEnglish
Pages (from-to)794-800
Number of pages7
JournalBJU International
Volume99
Issue number4
DOIs
Publication statusPublished - Apr 2007

Fingerprint

Nomograms
Prostatic Neoplasms
Logistic Models
Regression Analysis

Keywords

  • Artificial neural network
  • Classification and regression tree
  • Look-up table
  • Nomogram
  • Outcome
  • Prediction tools
  • Prostate cancer
  • Risk group stratification

ASJC Scopus subject areas

  • Urology

Cite this

A critical appraisal of logistic regression-based nomograms, artificial neural networks, classification and regression-tree models, look-up tables and risk-group stratification models for prostate cancer. / Chun, Felix K H; Karakiewicz, Pierre I.; Briganti, Alberto; Walz, Jochen; Kattan, Michael W.; Huland, Hartwig; Graefen, Markus.

In: BJU International, Vol. 99, No. 4, 04.2007, p. 794-800.

Research output: Contribution to journalArticle

@article{71562c6fc27a494eab478916088777e4,
title = "A critical appraisal of logistic regression-based nomograms, artificial neural networks, classification and regression-tree models, look-up tables and risk-group stratification models for prostate cancer",
abstract = "OBJECTIVE: To evaluate several methods of predicting prostate cancer-related outcomes, i.e. nomograms, look-up tables, artificial neural networks (ANN), classification and regression tree (CART) analyses and risk-group stratification (RGS) models, all of which represent valid alternatives. METHODS: We present four direct comparisons, where a nomogram was compared to either an ANN, a look-up table, a CART model or a RGS model. In all comparisons we assessed the predictive accuracy and performance characteristics of both models. RESULTS: Nomograms have several advantages over ANN, look-up tables, CART and RGS models, the most fundamental being a higher predictive accuracy and better performance characteristics. CONCLUSION: These results suggest that nomograms are more accurate and have better performance characteristics than their alternatives. However, ANN, look-up tables, CART analyses and RGS models all rely on methodologically sound and valid alternatives, which should not be abandoned.",
keywords = "Artificial neural network, Classification and regression tree, Look-up table, Nomogram, Outcome, Prediction tools, Prostate cancer, Risk group stratification",
author = "Chun, {Felix K H} and Karakiewicz, {Pierre I.} and Alberto Briganti and Jochen Walz and Kattan, {Michael W.} and Hartwig Huland and Markus Graefen",
year = "2007",
month = "4",
doi = "10.1111/j.1464-410X.2006.06694.x",
language = "English",
volume = "99",
pages = "794--800",
journal = "BJU International",
issn = "1464-4096",
publisher = "Wiley-Blackwell Publishing Ltd",
number = "4",

}

TY - JOUR

T1 - A critical appraisal of logistic regression-based nomograms, artificial neural networks, classification and regression-tree models, look-up tables and risk-group stratification models for prostate cancer

AU - Chun, Felix K H

AU - Karakiewicz, Pierre I.

AU - Briganti, Alberto

AU - Walz, Jochen

AU - Kattan, Michael W.

AU - Huland, Hartwig

AU - Graefen, Markus

PY - 2007/4

Y1 - 2007/4

N2 - OBJECTIVE: To evaluate several methods of predicting prostate cancer-related outcomes, i.e. nomograms, look-up tables, artificial neural networks (ANN), classification and regression tree (CART) analyses and risk-group stratification (RGS) models, all of which represent valid alternatives. METHODS: We present four direct comparisons, where a nomogram was compared to either an ANN, a look-up table, a CART model or a RGS model. In all comparisons we assessed the predictive accuracy and performance characteristics of both models. RESULTS: Nomograms have several advantages over ANN, look-up tables, CART and RGS models, the most fundamental being a higher predictive accuracy and better performance characteristics. CONCLUSION: These results suggest that nomograms are more accurate and have better performance characteristics than their alternatives. However, ANN, look-up tables, CART analyses and RGS models all rely on methodologically sound and valid alternatives, which should not be abandoned.

AB - OBJECTIVE: To evaluate several methods of predicting prostate cancer-related outcomes, i.e. nomograms, look-up tables, artificial neural networks (ANN), classification and regression tree (CART) analyses and risk-group stratification (RGS) models, all of which represent valid alternatives. METHODS: We present four direct comparisons, where a nomogram was compared to either an ANN, a look-up table, a CART model or a RGS model. In all comparisons we assessed the predictive accuracy and performance characteristics of both models. RESULTS: Nomograms have several advantages over ANN, look-up tables, CART and RGS models, the most fundamental being a higher predictive accuracy and better performance characteristics. CONCLUSION: These results suggest that nomograms are more accurate and have better performance characteristics than their alternatives. However, ANN, look-up tables, CART analyses and RGS models all rely on methodologically sound and valid alternatives, which should not be abandoned.

KW - Artificial neural network

KW - Classification and regression tree

KW - Look-up table

KW - Nomogram

KW - Outcome

KW - Prediction tools

KW - Prostate cancer

KW - Risk group stratification

UR - http://www.scopus.com/inward/record.url?scp=33947317768&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33947317768&partnerID=8YFLogxK

U2 - 10.1111/j.1464-410X.2006.06694.x

DO - 10.1111/j.1464-410X.2006.06694.x

M3 - Article

C2 - 17378842

AN - SCOPUS:33947317768

VL - 99

SP - 794

EP - 800

JO - BJU International

JF - BJU International

SN - 1464-4096

IS - 4

ER -