Multi-Criterion Mammographic Risk Analysis Supported with Multi-Label Fuzzy-Rough Feature Selection

Yanpeng Qu; Guanli Yue; Changjing Shang; Longzhi Yang; Reyer Zwiggelaar; Qiang Shen

doi:10.1016/j.artmed.2019.101722

Multi-Criterion Mammographic Risk Analysis Supported with Multi-Label Fuzzy-Rough Feature Selection

Yanpeng Qu, Guanli Yue, Changjing Shang, Longzhi Yang, Reyer Zwiggelaar, Qiang Shen

Cyfrifiadureg

Dalian Maritime University

Allbwn ymchwil: Cyfraniad at gyfnodolyn › Erthygl › adolygiad gan gymheiriaid

17 Dyfyniadau(SciVal)

166 Wedi eu Llwytho i Lawr (Pure)

Crynodeb

Context and background
Breast cancer is one of the most common diseases threatening the human lives globally, requiring effective and early risk analysis for which learning classifiers supported with automated feature selection offer a potential robust solution.

Motivation
Computer aided risk analysis of breast cancer typically works with a set of extracted mammographic features which may contain significant redundancy and noise, thereby requiring technical developments to improve runtime performance in both computational efficiency and classification accuracy.

Hypothesis
Use of advanced feature selection methods based on multiple diagnosis criteria may lead to improved results for mammographic risk analysis.

Methods
An approach for multi-criterion based mammographic risk analysis is proposed, by adapting the recently developed multi-label fuzzy-rough feature selection mechanism.

Results
A system for multi-criterion mammographic risk analysis is implemented with the aid of multi-label fuzzy-rough feature selection and its performance is positively verified experimentally, in comparison with representative popular mechanisms.

Conclusions
The novel approach for mammographic risk analysis based on multiple criteria helps improve classification accuracy using selected informative features, without suffering from the redundancy caused by such complex criteria, with the implemented system demonstrating practical efficacy.

Iaith wreiddiol	Saesneg
Rhif yr erthygl	101722
Cyfnodolyn	Artificial Intelligence in Medicine
Cyfrol	100
Dynodwyr Gwrthrych Digidol (DOIs)	https://doi.org/10.1016/j.artmed.2019.101722
Statws	Cyhoeddwyd - 25 Medi 2019

NDC y CU

Mae’r allbwn hwn yn cyfrannu at y Nod(au) Datblygu Cynaliadwy canlynol

Mynediad at Ddogfen

10.1016/j.artmed.2019.101722Trwydded: CC BY-NC-ND

Multi-criterion mammographic risk analysis supported with multi-label fuzzy-rough feature selectionFersiwn derfynol wedi’i chyhoeddi, 2.35 MBTrwydded: CC BY-NC-ND

Cysylltiad parhaol

Dolen barhaus

Ffeiliau eraill a chysylltiadau

Dolen i’r cyhoeddiad yn Scopus

1 Wedi Gorffen

Translating uncertainty computing techniques to computer aided diagnosis
Zwiggelaar, R.
The Royal Society
31 Maw 2017 → 30 Maw 2019
Prosiect: Ymchwil a ariannwyd yn allanol

Dyfynnu hyn

@article{0a4e592af94f4d5897a85ec51f4972fa,

title = "Multi-Criterion Mammographic Risk Analysis Supported with Multi-Label Fuzzy-Rough Feature Selection",

abstract = "Context and backgroundBreast cancer is one of the most common diseases threatening the human lives globally, requiring effective and early risk analysis for which learning classifiers supported with automated feature selection offer a potential robust solution.MotivationComputer aided risk analysis of breast cancer typically works with a set of extracted mammographic features which may contain significant redundancy and noise, thereby requiring technical developments to improve runtime performance in both computational efficiency and classification accuracy.HypothesisUse of advanced feature selection methods based on multiple diagnosis criteria may lead to improved results for mammographic risk analysis.MethodsAn approach for multi-criterion based mammographic risk analysis is proposed, by adapting the recently developed multi-label fuzzy-rough feature selection mechanism.ResultsA system for multi-criterion mammographic risk analysis is implemented with the aid of multi-label fuzzy-rough feature selection and its performance is positively verified experimentally, in comparison with representative popular mechanisms.ConclusionsThe novel approach for mammographic risk analysis based on multiple criteria helps improve classification accuracy using selected informative features, without suffering from the redundancy caused by such complex criteria, with the implemented system demonstrating practical efficacy.",

keywords = "learning classifiers, feature selection, multi criteria, multiple labels, fuzzy-rough dependency, mammographic risk, Feature selection, Multiple labels, Multiple criteria, Fuzzy-rough dependency, Learning classifiers, Mammographic risk, Humans, Machine Learning, Mammography/methods, Diagnosis, Computer-Assisted/methods, Breast Neoplasms/diagnosis, Algorithms, Female, Image Interpretation, Computer-Assisted/methods, Risk Assessment/methods, Fuzzy Logic",

author = "Yanpeng Qu and Guanli Yue and Changjing Shang and Longzhi Yang and Reyer Zwiggelaar and Qiang Shen",

note = "Funding Information: This research is jointly supported by the National Natural Science Foundation of China (No. 61502068 ), the Royal Society International Exchanges Cost Share Award with NSFC (No. IE160875 ), the Innovation Support Plan for Dalian High-level Talents (No. 2018RQ70 ), and a S{\^e}r Cymru II COFUND Fellowship, UK . The authors would like to thank the anonymous referees for their constructive comments which have been very helpful in revising this work. Publisher Copyright: {\textcopyright} 2019 The Authors",

year = "2019",

month = sep,

day = "25",

doi = "10.1016/j.artmed.2019.101722",

language = "English",

volume = "100",

journal = "Artificial Intelligence in Medicine",

issn = "0933-3657",

publisher = "Elsevier",

}

TY - JOUR

T1 - Multi-Criterion Mammographic Risk Analysis Supported with Multi-Label Fuzzy-Rough Feature Selection

AU - Qu, Yanpeng

AU - Yue, Guanli

AU - Shang, Changjing

AU - Yang, Longzhi

AU - Zwiggelaar, Reyer

AU - Shen, Qiang

N1 - Funding Information: This research is jointly supported by the National Natural Science Foundation of China (No. 61502068 ), the Royal Society International Exchanges Cost Share Award with NSFC (No. IE160875 ), the Innovation Support Plan for Dalian High-level Talents (No. 2018RQ70 ), and a Sêr Cymru II COFUND Fellowship, UK . The authors would like to thank the anonymous referees for their constructive comments which have been very helpful in revising this work. Publisher Copyright: © 2019 The Authors

PY - 2019/9/25

Y1 - 2019/9/25

N2 - Context and backgroundBreast cancer is one of the most common diseases threatening the human lives globally, requiring effective and early risk analysis for which learning classifiers supported with automated feature selection offer a potential robust solution.MotivationComputer aided risk analysis of breast cancer typically works with a set of extracted mammographic features which may contain significant redundancy and noise, thereby requiring technical developments to improve runtime performance in both computational efficiency and classification accuracy.HypothesisUse of advanced feature selection methods based on multiple diagnosis criteria may lead to improved results for mammographic risk analysis.MethodsAn approach for multi-criterion based mammographic risk analysis is proposed, by adapting the recently developed multi-label fuzzy-rough feature selection mechanism.ResultsA system for multi-criterion mammographic risk analysis is implemented with the aid of multi-label fuzzy-rough feature selection and its performance is positively verified experimentally, in comparison with representative popular mechanisms.ConclusionsThe novel approach for mammographic risk analysis based on multiple criteria helps improve classification accuracy using selected informative features, without suffering from the redundancy caused by such complex criteria, with the implemented system demonstrating practical efficacy.

AB - Context and backgroundBreast cancer is one of the most common diseases threatening the human lives globally, requiring effective and early risk analysis for which learning classifiers supported with automated feature selection offer a potential robust solution.MotivationComputer aided risk analysis of breast cancer typically works with a set of extracted mammographic features which may contain significant redundancy and noise, thereby requiring technical developments to improve runtime performance in both computational efficiency and classification accuracy.HypothesisUse of advanced feature selection methods based on multiple diagnosis criteria may lead to improved results for mammographic risk analysis.MethodsAn approach for multi-criterion based mammographic risk analysis is proposed, by adapting the recently developed multi-label fuzzy-rough feature selection mechanism.ResultsA system for multi-criterion mammographic risk analysis is implemented with the aid of multi-label fuzzy-rough feature selection and its performance is positively verified experimentally, in comparison with representative popular mechanisms.ConclusionsThe novel approach for mammographic risk analysis based on multiple criteria helps improve classification accuracy using selected informative features, without suffering from the redundancy caused by such complex criteria, with the implemented system demonstrating practical efficacy.

KW - learning classifiers

KW - feature selection

KW - multi criteria

KW - multiple labels

KW - fuzzy-rough dependency

KW - mammographic risk

KW - Feature selection

KW - Multiple labels

KW - Multiple criteria

KW - Fuzzy-rough dependency

KW - Learning classifiers

KW - Mammographic risk

KW - Humans

KW - Machine Learning

KW - Mammography/methods

KW - Diagnosis, Computer-Assisted/methods

KW - Breast Neoplasms/diagnosis

KW - Algorithms

KW - Female

KW - Image Interpretation, Computer-Assisted/methods

KW - Risk Assessment/methods

KW - Fuzzy Logic

UR - http://www.scopus.com/inward/record.url?scp=85072570508&partnerID=8YFLogxK

U2 - 10.1016/j.artmed.2019.101722

DO - 10.1016/j.artmed.2019.101722

M3 - Article

C2 - 31607343

SN - 0933-3657

VL - 100

JO - Artificial Intelligence in Medicine

JF - Artificial Intelligence in Medicine

M1 - 101722

ER -

Multi-Criterion Mammographic Risk Analysis Supported with Multi-Label Fuzzy-Rough Feature Selection

Crynodeb

NDC y CU

Mynediad at Ddogfen

Cysylltiad parhaol

Ffeiliau eraill a chysylltiadau

Ôl bys

Prosiectau

Translating uncertainty computing techniques to computer aided diagnosis

Dyfynnu hyn