Disease-free survival assessment by artificial neural networks for hepatocellular carcinoma patients after radiofrequency ablation

Chiueng Fang Wu, Yu Jen Wu, Po Chin Liang, Chih Horng Wu, Shinn Forng Peng, Hung Wen Chiu

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Background: Radiofrequency ablation (RFA) provides an effective treatment for patients who exhibit early hepatocellular carcinoma (HCC) stages or are waiting for liver transplantation. It is important to assess patients after RFA. The goal of this study was to build artificial neural network models with HCC-related variables to predict the 1-year and 2-year disease-free survival (DFS) of HCC patients receiving RFA treatments. Methods: This study was a retrospective study that tracked HCC patients who received computer tomography-guided percutaneous RFA between January 2009 and April 2012. The numbers of total patients with 1-year and 2-year DFS were 252 and 179, respectively. A total of 15 HCC clinical variables were collected for the construction of artificial neural network models for DFS prediction. Internal validation and validation conducted using simulated prospective data were performed. Results: The results showed that the model with 15 inputs showed better performance compared with the models including only significant features. Parameters for performance assessment of 1-year DFS prediction were as follows: accuracy 85.0% (70.0%), sensitivity 75.0% (63.3%), specificity 87.5% (71.8%), and area under the curve 0.84 (0.77) for internal validation (simulated prospective validation). For 2-year DFS prediction, the values of accuracy, sensitivity, specificity, and area under the curve were 67.9% (63.9%), 50.0% (56.3%), 85.7% (70.0%), and 0.75 (0.72), respectively, for internal validation (simulated prospective validation). Conclusion: This study revealed that the proposed artificial neural network models constructed with 15 clinical HCC relevant features could achieve an acceptable prediction performance for DFS. Such models can support clinical physicians to deal with clinical decision-making processes on the prognosis of HCC patients receiving RFA treatments.

Original languageEnglish
Pages (from-to)765-773
JournalJournal of the Formosan Medical Association
DOIs
Publication statusPublished - Oct 2017

Fingerprint

Disease-Free Survival
Hepatocellular Carcinoma
Neural Networks (Computer)
Area Under Curve
Liver Transplantation
Therapeutics
Retrospective Studies
Tomography
Physicians
Sensitivity and Specificity

Keywords

  • Artificial neural network
  • Disease-free survival
  • Hepatocellular carcinoma
  • Medical decision support
  • Radiofrequency ablation

ASJC Scopus subject areas

  • Medicine(all)

Cite this

Disease-free survival assessment by artificial neural networks for hepatocellular carcinoma patients after radiofrequency ablation. / Wu, Chiueng Fang; Wu, Yu Jen; Liang, Po Chin; Wu, Chih Horng; Peng, Shinn Forng; Chiu, Hung Wen.

In: Journal of the Formosan Medical Association, 10.2017, p. 765-773.

Research output: Contribution to journalArticle

@article{d5d2314d75274beeb021b59215c8dc99,
title = "Disease-free survival assessment by artificial neural networks for hepatocellular carcinoma patients after radiofrequency ablation",
abstract = "Background: Radiofrequency ablation (RFA) provides an effective treatment for patients who exhibit early hepatocellular carcinoma (HCC) stages or are waiting for liver transplantation. It is important to assess patients after RFA. The goal of this study was to build artificial neural network models with HCC-related variables to predict the 1-year and 2-year disease-free survival (DFS) of HCC patients receiving RFA treatments. Methods: This study was a retrospective study that tracked HCC patients who received computer tomography-guided percutaneous RFA between January 2009 and April 2012. The numbers of total patients with 1-year and 2-year DFS were 252 and 179, respectively. A total of 15 HCC clinical variables were collected for the construction of artificial neural network models for DFS prediction. Internal validation and validation conducted using simulated prospective data were performed. Results: The results showed that the model with 15 inputs showed better performance compared with the models including only significant features. Parameters for performance assessment of 1-year DFS prediction were as follows: accuracy 85.0{\%} (70.0{\%}), sensitivity 75.0{\%} (63.3{\%}), specificity 87.5{\%} (71.8{\%}), and area under the curve 0.84 (0.77) for internal validation (simulated prospective validation). For 2-year DFS prediction, the values of accuracy, sensitivity, specificity, and area under the curve were 67.9{\%} (63.9{\%}), 50.0{\%} (56.3{\%}), 85.7{\%} (70.0{\%}), and 0.75 (0.72), respectively, for internal validation (simulated prospective validation). Conclusion: This study revealed that the proposed artificial neural network models constructed with 15 clinical HCC relevant features could achieve an acceptable prediction performance for DFS. Such models can support clinical physicians to deal with clinical decision-making processes on the prognosis of HCC patients receiving RFA treatments.",
keywords = "Artificial neural network, Disease-free survival, Hepatocellular carcinoma, Medical decision support, Radiofrequency ablation",
author = "Wu, {Chiueng Fang} and Wu, {Yu Jen} and Liang, {Po Chin} and Wu, {Chih Horng} and Peng, {Shinn Forng} and Chiu, {Hung Wen}",
year = "2017",
month = "10",
doi = "10.1016/j.jfma.2016.12.006",
language = "English",
pages = "765--773",
journal = "Journal of the Formosan Medical Association",
issn = "0929-6646",
publisher = "Elsevier Science Publishers B.V.",

}

TY - JOUR

T1 - Disease-free survival assessment by artificial neural networks for hepatocellular carcinoma patients after radiofrequency ablation

AU - Wu, Chiueng Fang

AU - Wu, Yu Jen

AU - Liang, Po Chin

AU - Wu, Chih Horng

AU - Peng, Shinn Forng

AU - Chiu, Hung Wen

PY - 2017/10

Y1 - 2017/10

N2 - Background: Radiofrequency ablation (RFA) provides an effective treatment for patients who exhibit early hepatocellular carcinoma (HCC) stages or are waiting for liver transplantation. It is important to assess patients after RFA. The goal of this study was to build artificial neural network models with HCC-related variables to predict the 1-year and 2-year disease-free survival (DFS) of HCC patients receiving RFA treatments. Methods: This study was a retrospective study that tracked HCC patients who received computer tomography-guided percutaneous RFA between January 2009 and April 2012. The numbers of total patients with 1-year and 2-year DFS were 252 and 179, respectively. A total of 15 HCC clinical variables were collected for the construction of artificial neural network models for DFS prediction. Internal validation and validation conducted using simulated prospective data were performed. Results: The results showed that the model with 15 inputs showed better performance compared with the models including only significant features. Parameters for performance assessment of 1-year DFS prediction were as follows: accuracy 85.0% (70.0%), sensitivity 75.0% (63.3%), specificity 87.5% (71.8%), and area under the curve 0.84 (0.77) for internal validation (simulated prospective validation). For 2-year DFS prediction, the values of accuracy, sensitivity, specificity, and area under the curve were 67.9% (63.9%), 50.0% (56.3%), 85.7% (70.0%), and 0.75 (0.72), respectively, for internal validation (simulated prospective validation). Conclusion: This study revealed that the proposed artificial neural network models constructed with 15 clinical HCC relevant features could achieve an acceptable prediction performance for DFS. Such models can support clinical physicians to deal with clinical decision-making processes on the prognosis of HCC patients receiving RFA treatments.

AB - Background: Radiofrequency ablation (RFA) provides an effective treatment for patients who exhibit early hepatocellular carcinoma (HCC) stages or are waiting for liver transplantation. It is important to assess patients after RFA. The goal of this study was to build artificial neural network models with HCC-related variables to predict the 1-year and 2-year disease-free survival (DFS) of HCC patients receiving RFA treatments. Methods: This study was a retrospective study that tracked HCC patients who received computer tomography-guided percutaneous RFA between January 2009 and April 2012. The numbers of total patients with 1-year and 2-year DFS were 252 and 179, respectively. A total of 15 HCC clinical variables were collected for the construction of artificial neural network models for DFS prediction. Internal validation and validation conducted using simulated prospective data were performed. Results: The results showed that the model with 15 inputs showed better performance compared with the models including only significant features. Parameters for performance assessment of 1-year DFS prediction were as follows: accuracy 85.0% (70.0%), sensitivity 75.0% (63.3%), specificity 87.5% (71.8%), and area under the curve 0.84 (0.77) for internal validation (simulated prospective validation). For 2-year DFS prediction, the values of accuracy, sensitivity, specificity, and area under the curve were 67.9% (63.9%), 50.0% (56.3%), 85.7% (70.0%), and 0.75 (0.72), respectively, for internal validation (simulated prospective validation). Conclusion: This study revealed that the proposed artificial neural network models constructed with 15 clinical HCC relevant features could achieve an acceptable prediction performance for DFS. Such models can support clinical physicians to deal with clinical decision-making processes on the prognosis of HCC patients receiving RFA treatments.

KW - Artificial neural network

KW - Disease-free survival

KW - Hepatocellular carcinoma

KW - Medical decision support

KW - Radiofrequency ablation

UR - http://www.scopus.com/inward/record.url?scp=85009808404&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85009808404&partnerID=8YFLogxK

U2 - 10.1016/j.jfma.2016.12.006

DO - 10.1016/j.jfma.2016.12.006

M3 - Article

AN - SCOPUS:85009808404

SP - 765

EP - 773

JO - Journal of the Formosan Medical Association

JF - Journal of the Formosan Medical Association

SN - 0929-6646

ER -