Deep learning-based survival prediction of oral cancer patients

Dong Wook Kim, Sanghoon Lee, Sunmo Kwon, Woong Nam, Inho Cha, Hyung Jun Kim

Research output: Contribution to journalArticle

Abstract

The Cox proportional hazards model commonly used to evaluate prognostic variables in survival of cancer patients may be too simplistic to properly predict a cancer patient’s outcome since it assumes that the outcome is a linear combination of covariates. In this retrospective study including 255 patients suitable for analysis who underwent surgical treatment in our department from 2000 to 2017, we applied a deep learning-based survival prediction method in oral squamous cell carcinoma (SCC) patients and validated its performance. Survival prediction using DeepSurv, a deep learning based-survival prediction algorithm, was compared with random survival forest (RSF) and the Cox proportional hazard model (CPH). DeepSurv showed the best performance among the three models, the c-index of the training and testing sets reaching 0.810 and 0.781, respectively, followed by RSF (0.770/0.764), and CPH (0.756/0.694). The performance of DeepSurv steadily improved with added features. Thus, deep learning-based survival prediction may improve prediction accuracy and guide clinicians both in choosing treatment options for better survival and in avoiding unnecessary treatments.

Original languageEnglish
Article number6994
JournalScientific reports
Volume9
Issue number1
DOIs
Publication statusPublished - 2019 Dec 1

Fingerprint

Mouth Neoplasms
Learning
Survival
Proportional Hazards Models
Squamous Cell Carcinoma
Neoplasms
Therapeutics
Retrospective Studies

All Science Journal Classification (ASJC) codes

  • General

Cite this

Kim, Dong Wook ; Lee, Sanghoon ; Kwon, Sunmo ; Nam, Woong ; Cha, Inho ; Kim, Hyung Jun. / Deep learning-based survival prediction of oral cancer patients. In: Scientific reports. 2019 ; Vol. 9, No. 1.
@article{0823868d55bb4e4ca08e10465e486684,
title = "Deep learning-based survival prediction of oral cancer patients",
abstract = "The Cox proportional hazards model commonly used to evaluate prognostic variables in survival of cancer patients may be too simplistic to properly predict a cancer patient’s outcome since it assumes that the outcome is a linear combination of covariates. In this retrospective study including 255 patients suitable for analysis who underwent surgical treatment in our department from 2000 to 2017, we applied a deep learning-based survival prediction method in oral squamous cell carcinoma (SCC) patients and validated its performance. Survival prediction using DeepSurv, a deep learning based-survival prediction algorithm, was compared with random survival forest (RSF) and the Cox proportional hazard model (CPH). DeepSurv showed the best performance among the three models, the c-index of the training and testing sets reaching 0.810 and 0.781, respectively, followed by RSF (0.770/0.764), and CPH (0.756/0.694). The performance of DeepSurv steadily improved with added features. Thus, deep learning-based survival prediction may improve prediction accuracy and guide clinicians both in choosing treatment options for better survival and in avoiding unnecessary treatments.",
author = "Kim, {Dong Wook} and Sanghoon Lee and Sunmo Kwon and Woong Nam and Inho Cha and Kim, {Hyung Jun}",
year = "2019",
month = "12",
day = "1",
doi = "10.1038/s41598-019-43372-7",
language = "English",
volume = "9",
journal = "Scientific Reports",
issn = "2045-2322",
publisher = "Nature Publishing Group",
number = "1",

}

Deep learning-based survival prediction of oral cancer patients. / Kim, Dong Wook; Lee, Sanghoon; Kwon, Sunmo; Nam, Woong; Cha, Inho; Kim, Hyung Jun.

In: Scientific reports, Vol. 9, No. 1, 6994, 01.12.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Deep learning-based survival prediction of oral cancer patients

AU - Kim, Dong Wook

AU - Lee, Sanghoon

AU - Kwon, Sunmo

AU - Nam, Woong

AU - Cha, Inho

AU - Kim, Hyung Jun

PY - 2019/12/1

Y1 - 2019/12/1

N2 - The Cox proportional hazards model commonly used to evaluate prognostic variables in survival of cancer patients may be too simplistic to properly predict a cancer patient’s outcome since it assumes that the outcome is a linear combination of covariates. In this retrospective study including 255 patients suitable for analysis who underwent surgical treatment in our department from 2000 to 2017, we applied a deep learning-based survival prediction method in oral squamous cell carcinoma (SCC) patients and validated its performance. Survival prediction using DeepSurv, a deep learning based-survival prediction algorithm, was compared with random survival forest (RSF) and the Cox proportional hazard model (CPH). DeepSurv showed the best performance among the three models, the c-index of the training and testing sets reaching 0.810 and 0.781, respectively, followed by RSF (0.770/0.764), and CPH (0.756/0.694). The performance of DeepSurv steadily improved with added features. Thus, deep learning-based survival prediction may improve prediction accuracy and guide clinicians both in choosing treatment options for better survival and in avoiding unnecessary treatments.

AB - The Cox proportional hazards model commonly used to evaluate prognostic variables in survival of cancer patients may be too simplistic to properly predict a cancer patient’s outcome since it assumes that the outcome is a linear combination of covariates. In this retrospective study including 255 patients suitable for analysis who underwent surgical treatment in our department from 2000 to 2017, we applied a deep learning-based survival prediction method in oral squamous cell carcinoma (SCC) patients and validated its performance. Survival prediction using DeepSurv, a deep learning based-survival prediction algorithm, was compared with random survival forest (RSF) and the Cox proportional hazard model (CPH). DeepSurv showed the best performance among the three models, the c-index of the training and testing sets reaching 0.810 and 0.781, respectively, followed by RSF (0.770/0.764), and CPH (0.756/0.694). The performance of DeepSurv steadily improved with added features. Thus, deep learning-based survival prediction may improve prediction accuracy and guide clinicians both in choosing treatment options for better survival and in avoiding unnecessary treatments.

UR - http://www.scopus.com/inward/record.url?scp=85065332145&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85065332145&partnerID=8YFLogxK

U2 - 10.1038/s41598-019-43372-7

DO - 10.1038/s41598-019-43372-7

M3 - Article

C2 - 31061433

AN - SCOPUS:85065332145

VL - 9

JO - Scientific Reports

JF - Scientific Reports

SN - 2045-2322

IS - 1

M1 - 6994

ER -