Machine learning techniques in cancer prognostic modeling and performance assessment

Yiyi Chen; Jess A. Millar

doi:10.1007/978-981-10-0126-0_13

Machine learning techniques in cancer prognostic modeling and performance assessment

Yiyi Chen, Jess A. Millar

School Of Public Health

Research output: Chapter in Book/Report/Conference proceeding › Chapter

3 Scopus citations

Abstract

Prognostic models for disease occurrence, tumor progression and survival are abundant for most types of cancers. Physicians and cancer patients are utilizing these models to make informed treatment decisions and corresponding arrangements. However, not all cancer prognostic models are built and validated rigorously. Some are more useful and reliable than others. In this chapter, we briefly introduce some popular machine learning methods for constructing cancer prognostic models, and discuss pros and cons of each. We also introduce the commonly used discrimination and calibration metrics for assessing predictive performance and validating the prognostic models. In the end, we outline several challenges of using prognostic models in the real world for clinical decision-making support, and propose related suggestions.

Original language	English (US)
Title of host publication	Frontiers of Biostatistical Methods and Applications in Clinical Oncology
Publisher	Springer Singapore
Pages	193-230
Number of pages	38
ISBN (Electronic)	9789811001260
ISBN (Print)	9789811001246
DOIs	https://doi.org/10.1007/978-981-10-0126-0_13
State	Published - Oct 3 2017

Keywords

Cancer prediction
Machine learning
Prognostic model
Validation

ASJC Scopus subject areas

General Medicine
General Mathematics
General Social Sciences

Access to Document

10.1007/978-981-10-0126-0_13

Cite this

@inbook{c837ab0283fd4d679e4d0b40f4c0c3db,

title = "Machine learning techniques in cancer prognostic modeling and performance assessment",

abstract = "Prognostic models for disease occurrence, tumor progression and survival are abundant for most types of cancers. Physicians and cancer patients are utilizing these models to make informed treatment decisions and corresponding arrangements. However, not all cancer prognostic models are built and validated rigorously. Some are more useful and reliable than others. In this chapter, we briefly introduce some popular machine learning methods for constructing cancer prognostic models, and discuss pros and cons of each. We also introduce the commonly used discrimination and calibration metrics for assessing predictive performance and validating the prognostic models. In the end, we outline several challenges of using prognostic models in the real world for clinical decision-making support, and propose related suggestions.",

keywords = "Cancer prediction, Machine learning, Prognostic model, Validation",

author = "Yiyi Chen and Millar, {Jess A.}",

year = "2017",

month = oct,

day = "3",

doi = "10.1007/978-981-10-0126-0_13",

language = "English (US)",

isbn = "9789811001246",

pages = "193--230",

booktitle = "Frontiers of Biostatistical Methods and Applications in Clinical Oncology",

publisher = "Springer Singapore",

}

TY - CHAP

T1 - Machine learning techniques in cancer prognostic modeling and performance assessment

AU - Chen, Yiyi

AU - Millar, Jess A.

PY - 2017/10/3

Y1 - 2017/10/3

N2 - Prognostic models for disease occurrence, tumor progression and survival are abundant for most types of cancers. Physicians and cancer patients are utilizing these models to make informed treatment decisions and corresponding arrangements. However, not all cancer prognostic models are built and validated rigorously. Some are more useful and reliable than others. In this chapter, we briefly introduce some popular machine learning methods for constructing cancer prognostic models, and discuss pros and cons of each. We also introduce the commonly used discrimination and calibration metrics for assessing predictive performance and validating the prognostic models. In the end, we outline several challenges of using prognostic models in the real world for clinical decision-making support, and propose related suggestions.

AB - Prognostic models for disease occurrence, tumor progression and survival are abundant for most types of cancers. Physicians and cancer patients are utilizing these models to make informed treatment decisions and corresponding arrangements. However, not all cancer prognostic models are built and validated rigorously. Some are more useful and reliable than others. In this chapter, we briefly introduce some popular machine learning methods for constructing cancer prognostic models, and discuss pros and cons of each. We also introduce the commonly used discrimination and calibration metrics for assessing predictive performance and validating the prognostic models. In the end, we outline several challenges of using prognostic models in the real world for clinical decision-making support, and propose related suggestions.

KW - Cancer prediction

KW - Machine learning

KW - Prognostic model

KW - Validation

UR - http://www.scopus.com/inward/record.url?scp=85042635895&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85042635895&partnerID=8YFLogxK

U2 - 10.1007/978-981-10-0126-0_13

DO - 10.1007/978-981-10-0126-0_13

M3 - Chapter

AN - SCOPUS:85042635895

SN - 9789811001246

SP - 193

EP - 230

BT - Frontiers of Biostatistical Methods and Applications in Clinical Oncology

PB - Springer Singapore

ER -

Machine learning techniques in cancer prognostic modeling and performance assessment

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this