Artificial neural network for predicting pathological stage of clinically localized prostate cancer in a Taiwanese population

Chih Wei Tsao, Ching Yu Liu, Tai Lung Cha, Sheng Tang Wu, Guang Huan Sun, Dah Shyong Yu, Hong I. Chen, Sun Yran Chang, Shih Chang Chen, Chien Yeh Hsu

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

Background: We developed an artificial neural network (ANN) model to predict prostate cancer pathological staging in patients prior to when they received radical prostatectomy as this is more effective than logistic regression (LR), or combined use of age, prostate-specific antigen (PSA), body mass index (BMI), digital rectal examination (DRE), trans-rectal ultrasound (TRUS), biopsy Gleason sum, and primary biopsy Gleason grade. Methods: Our study evaluated 299 patients undergoing retro-pubic radical prostatectomy or robotic-assisted laparoscopic radical prostatectomy surgical procedures with pelvic lymph node dissection. The results were intended to predict the pathological stage of prostate cancer (T2 or T3) after radical surgery. The predictive ability of ANN was compared with LR and validation of the 2007 Partin Tables was estimated by the areas under the receiving operating characteristic curve (AUCs). Results: Of the 299 patients we evaluated, 109 (36.45%) displayed prostate cancer with extra-capsular extension (ECE), and 190 (63.55%) displayed organ-confined disease (OCD). LR analysis showed that only PSA and BMI were statistically significant predictors of prostate cancer with capsule invasion. Overall, ANN outperformed LR significantly (0.795 ± 0.023 versus 0.746 ± 0.025, p = 0.016). Validation using the current Partin Tables for the participants of our study was assessed, and the predictive capacity of AUC for OCD was 0.695. Conclusion: ANN was superior to LR at predicting OCD in prostate cancer. Compared with the validation of current Partin Tables for the Taiwanese population, the ANN model resulted in larger AUCs and more accurate prediction of the pathologic stage of prostate cancer.

Original languageEnglish
JournalJournal of the Chinese Medical Association
DOIs
Publication statusAccepted/In press - 2014

Fingerprint

Prostatic Neoplasms
Logistic Models
Nomograms
Prostatectomy
Population
Area Under Curve
Neural Networks (Computer)
Prostate-Specific Antigen
Body Mass Index
Biopsy
Digital Rectal Examination
Neoplasm Staging
Robotics
Lymph Node Excision
Capsules
Regression Analysis

Keywords

  • artificial neural network
  • capsule invasion
  • Partin Tables
  • prostate neoplasm

ASJC Scopus subject areas

  • Medicine(all)

Cite this

Artificial neural network for predicting pathological stage of clinically localized prostate cancer in a Taiwanese population. / Tsao, Chih Wei; Liu, Ching Yu; Cha, Tai Lung; Wu, Sheng Tang; Sun, Guang Huan; Yu, Dah Shyong; Chen, Hong I.; Chang, Sun Yran; Chen, Shih Chang; Hsu, Chien Yeh.

In: Journal of the Chinese Medical Association, 2014.

Research output: Contribution to journalArticle

Tsao, Chih Wei ; Liu, Ching Yu ; Cha, Tai Lung ; Wu, Sheng Tang ; Sun, Guang Huan ; Yu, Dah Shyong ; Chen, Hong I. ; Chang, Sun Yran ; Chen, Shih Chang ; Hsu, Chien Yeh. / Artificial neural network for predicting pathological stage of clinically localized prostate cancer in a Taiwanese population. In: Journal of the Chinese Medical Association. 2014.
@article{0caea613c5664aa2b690f547caf8361d,
title = "Artificial neural network for predicting pathological stage of clinically localized prostate cancer in a Taiwanese population",
abstract = "Background: We developed an artificial neural network (ANN) model to predict prostate cancer pathological staging in patients prior to when they received radical prostatectomy as this is more effective than logistic regression (LR), or combined use of age, prostate-specific antigen (PSA), body mass index (BMI), digital rectal examination (DRE), trans-rectal ultrasound (TRUS), biopsy Gleason sum, and primary biopsy Gleason grade. Methods: Our study evaluated 299 patients undergoing retro-pubic radical prostatectomy or robotic-assisted laparoscopic radical prostatectomy surgical procedures with pelvic lymph node dissection. The results were intended to predict the pathological stage of prostate cancer (T2 or T3) after radical surgery. The predictive ability of ANN was compared with LR and validation of the 2007 Partin Tables was estimated by the areas under the receiving operating characteristic curve (AUCs). Results: Of the 299 patients we evaluated, 109 (36.45{\%}) displayed prostate cancer with extra-capsular extension (ECE), and 190 (63.55{\%}) displayed organ-confined disease (OCD). LR analysis showed that only PSA and BMI were statistically significant predictors of prostate cancer with capsule invasion. Overall, ANN outperformed LR significantly (0.795 ± 0.023 versus 0.746 ± 0.025, p = 0.016). Validation using the current Partin Tables for the participants of our study was assessed, and the predictive capacity of AUC for OCD was 0.695. Conclusion: ANN was superior to LR at predicting OCD in prostate cancer. Compared with the validation of current Partin Tables for the Taiwanese population, the ANN model resulted in larger AUCs and more accurate prediction of the pathologic stage of prostate cancer.",
keywords = "artificial neural network, capsule invasion, Partin Tables, prostate neoplasm",
author = "Tsao, {Chih Wei} and Liu, {Ching Yu} and Cha, {Tai Lung} and Wu, {Sheng Tang} and Sun, {Guang Huan} and Yu, {Dah Shyong} and Chen, {Hong I.} and Chang, {Sun Yran} and Chen, {Shih Chang} and Hsu, {Chien Yeh}",
year = "2014",
doi = "10.1016/j.jcma.2014.06.014",
language = "English",
journal = "Journal of the Chinese Medical Association",
issn = "1726-4901",
publisher = "Elsevier Taiwan LLC",

}

TY - JOUR

T1 - Artificial neural network for predicting pathological stage of clinically localized prostate cancer in a Taiwanese population

AU - Tsao, Chih Wei

AU - Liu, Ching Yu

AU - Cha, Tai Lung

AU - Wu, Sheng Tang

AU - Sun, Guang Huan

AU - Yu, Dah Shyong

AU - Chen, Hong I.

AU - Chang, Sun Yran

AU - Chen, Shih Chang

AU - Hsu, Chien Yeh

PY - 2014

Y1 - 2014

N2 - Background: We developed an artificial neural network (ANN) model to predict prostate cancer pathological staging in patients prior to when they received radical prostatectomy as this is more effective than logistic regression (LR), or combined use of age, prostate-specific antigen (PSA), body mass index (BMI), digital rectal examination (DRE), trans-rectal ultrasound (TRUS), biopsy Gleason sum, and primary biopsy Gleason grade. Methods: Our study evaluated 299 patients undergoing retro-pubic radical prostatectomy or robotic-assisted laparoscopic radical prostatectomy surgical procedures with pelvic lymph node dissection. The results were intended to predict the pathological stage of prostate cancer (T2 or T3) after radical surgery. The predictive ability of ANN was compared with LR and validation of the 2007 Partin Tables was estimated by the areas under the receiving operating characteristic curve (AUCs). Results: Of the 299 patients we evaluated, 109 (36.45%) displayed prostate cancer with extra-capsular extension (ECE), and 190 (63.55%) displayed organ-confined disease (OCD). LR analysis showed that only PSA and BMI were statistically significant predictors of prostate cancer with capsule invasion. Overall, ANN outperformed LR significantly (0.795 ± 0.023 versus 0.746 ± 0.025, p = 0.016). Validation using the current Partin Tables for the participants of our study was assessed, and the predictive capacity of AUC for OCD was 0.695. Conclusion: ANN was superior to LR at predicting OCD in prostate cancer. Compared with the validation of current Partin Tables for the Taiwanese population, the ANN model resulted in larger AUCs and more accurate prediction of the pathologic stage of prostate cancer.

AB - Background: We developed an artificial neural network (ANN) model to predict prostate cancer pathological staging in patients prior to when they received radical prostatectomy as this is more effective than logistic regression (LR), or combined use of age, prostate-specific antigen (PSA), body mass index (BMI), digital rectal examination (DRE), trans-rectal ultrasound (TRUS), biopsy Gleason sum, and primary biopsy Gleason grade. Methods: Our study evaluated 299 patients undergoing retro-pubic radical prostatectomy or robotic-assisted laparoscopic radical prostatectomy surgical procedures with pelvic lymph node dissection. The results were intended to predict the pathological stage of prostate cancer (T2 or T3) after radical surgery. The predictive ability of ANN was compared with LR and validation of the 2007 Partin Tables was estimated by the areas under the receiving operating characteristic curve (AUCs). Results: Of the 299 patients we evaluated, 109 (36.45%) displayed prostate cancer with extra-capsular extension (ECE), and 190 (63.55%) displayed organ-confined disease (OCD). LR analysis showed that only PSA and BMI were statistically significant predictors of prostate cancer with capsule invasion. Overall, ANN outperformed LR significantly (0.795 ± 0.023 versus 0.746 ± 0.025, p = 0.016). Validation using the current Partin Tables for the participants of our study was assessed, and the predictive capacity of AUC for OCD was 0.695. Conclusion: ANN was superior to LR at predicting OCD in prostate cancer. Compared with the validation of current Partin Tables for the Taiwanese population, the ANN model resulted in larger AUCs and more accurate prediction of the pathologic stage of prostate cancer.

KW - artificial neural network

KW - capsule invasion

KW - Partin Tables

KW - prostate neoplasm

UR - http://www.scopus.com/inward/record.url?scp=84906377181&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84906377181&partnerID=8YFLogxK

U2 - 10.1016/j.jcma.2014.06.014

DO - 10.1016/j.jcma.2014.06.014

M3 - Article

JO - Journal of the Chinese Medical Association

JF - Journal of the Chinese Medical Association

SN - 1726-4901

ER -