CSE database: extended annotations and new recommendations for
ECG software testing

J 2017

CSE database: extended annotations and new recommendations for ECG software testing

SMÍŠEK, R., Lucie MARŠÁNOVÁ, Andrea NĚMCOVÁ, Martin VÍTEK, Jiří KOZUMPLÍK et. al.

Basic information

Original name

CSE database: extended annotations and new recommendations for ECG software testing

Authors

SMÍŠEK, R. (203 Czech Republic), Lucie MARŠÁNOVÁ (203 Czech Republic), Andrea NĚMCOVÁ (203 Czech Republic), Martin VÍTEK (203 Czech Republic), Jiří KOZUMPLÍK (203 Czech Republic) and Marie NOVÁKOVÁ (203 Czech Republic, guarantor, belonging to the institution)

Edition

Medical and Biological Engineering and Computing, Heidelberg, Springer, 2017, 0140-0118

Other information

Language

English

Type of outcome

Článek v odborném periodiku

Field of Study

20601 Medical engineering

Country of publisher

Germany

Confidentiality degree

není předmětem státního či obchodního tajemství

Impact factor

Impact factor: 1.971

RIV identification code

RIV/00216224:14110/17:00094605

Organization unit

Faculty of Medicine

DOI

http://dx.doi.org/10.1007/s11517-016-1607-5

UT WoS

000407310300027

Keywords in English

Annotation of ECG record; CSE database; ECG; ECG classification; Recommendations; Software testing

Abstract

V originále

Nowadays, cardiovascular diseases represent the most common cause of death in western countries. Among various examination techniques, electrocardiography (ECG) is still a highly valuable tool used for the diagnosis of many cardiovascular disorders. In order to diagnose a person based on ECG, cardiologists can use automatic diagnostic algorithms. Research in this area is still necessary. In order to compare various algorithms correctly, it is necessary to test them on standard annotated databases, such as the Common Standards for Quantitative Electrocardiography (CSE) database. According to Scopus, the CSE database is the second most cited standard database. There were two main objectives in this work. First, new diagnoses were added to the CSE database, which extended its original annotations. Second, new recommendations for diagnostic software quality estimation were established. The ECG recordings were diagnosed by five new cardiologists independently, and in total, 59 different diagnoses were found. Such a large number of diagnoses is unique, even in terms of standard databases. Based on the cardiologists’ diagnoses, a four-round consensus (4R consensus) was established. Such a 4R consensus means a correct final diagnosis, which should ideally be the output of any tested classification software. The accuracy of the cardiologists’ diagnoses compared with the 4R consensus was the basis for the establishment of accuracy recommendations. The accuracy was determined in terms of sensitivity = 79.20–86.81%, positive predictive value = 79.10–87.11%, and the Jaccard coefficient = 72.21–81.14%, respectively. Within these ranges, the accuracy of the software is comparable with the accuracy of cardiologists. The accuracy quantification of the correct classification is unique. Diagnostic software developers can objectively evaluate the success of their algorithm and promote its further development. The annotations and recommendations proposed in this work will allow for faster development and testing of classification software. As a result, this might facilitate cardiologists’ work and lead to faster diagnoses and earlier treatment.

Links

GAP102/12/2034, research and development project

Name: Analýza vztahu mezi elektrickými ději a průtokem krve u srdečních komor

Investor: Czech Science Foundation

Citovat

SMÍŠEK, R., Lucie MARŠÁNOVÁ, Andrea NĚMCOVÁ, Martin VÍTEK, Jiří KOZUMPLÍK and Marie NOVÁKOVÁ. CSE database: extended annotations and new recommendations for ECG software testing. Medical and Biological Engineering and Computing. Heidelberg: Springer, 2017, vol. 55, No 8, p. 1473-1482. ISSN 0140-0118. Available from: https://dx.doi.org/10.1007/s11517-016-1607-5.

@article{1367533,
   author = {Smíšek, R. and Maršánová, Lucie and Němcová, Andrea and Vítek, Martin and Kozumplík, Jiří and Nováková, Marie},
   article_location = {Heidelberg},
   article_number = {8},
   doi = {http://dx.doi.org/10.1007/s11517-016-1607-5},
   keywords = {Annotation of ECG record; CSE database; ECG; ECG classification; Recommendations; Software testing},
   language = {eng},
   issn = {0140-0118},
   journal = {Medical and Biological Engineering and Computing},
   title = {CSE database: extended annotations and new recommendations for ECG software testing},
   volume = {55},
   year = {2017}
}

TY  - JOUR
ID  - 1367533
AU  - Smíšek, R. - Maršánová, Lucie - Němcová, Andrea - Vítek, Martin - Kozumplík, Jiří - Nováková, Marie
PY  - 2017
TI  - CSE database: extended annotations and new recommendations for ECG software testing
JF  - Medical and Biological Engineering and Computing
VL  - 55
IS  - 8
SP  - 1473-1482
EP  - 1473-1482
PB  - Springer
SN  - 01400118
KW  - Annotation of ECG record
KW  - CSE database
KW  - ECG
KW  - ECG classification
KW  - Recommendations
KW  - Software testing
N2  - Nowadays, cardiovascular diseases represent the most common cause of death in western countries. Among various examination techniques, electrocardiography (ECG) is still a highly valuable tool used for the diagnosis of many cardiovascular disorders. In order to diagnose a person based on ECG, cardiologists can use automatic diagnostic algorithms. Research in this area is still necessary. In order to compare various algorithms correctly, it is necessary to test them on standard annotated databases, such as the Common Standards for Quantitative Electrocardiography (CSE) database. According to Scopus, the CSE database is the second most cited standard database. There were two main objectives in this work. First, new diagnoses were added to the CSE database, which extended its original annotations. Second, new recommendations for diagnostic software quality estimation were established. The ECG recordings were diagnosed by five new cardiologists independently, and in total, 59 different diagnoses were found. Such a large number of diagnoses is unique, even in terms of standard databases. Based on the cardiologists’ diagnoses, a four-round consensus (4R consensus) was established. Such a 4R consensus means a correct final diagnosis, which should ideally be the output of any tested classification software. The accuracy of the cardiologists’ diagnoses compared with the 4R consensus was the basis for the establishment of accuracy recommendations. The accuracy was determined in terms of sensitivity = 79.20–86.81%, positive predictive value = 79.10–87.11%, and the Jaccard coefficient = 72.21–81.14%, respectively. Within these ranges, the accuracy of the software is comparable with the accuracy of cardiologists. The accuracy quantification of the correct classification is unique. Diagnostic software developers can objectively evaluate the success of their algorithm and promote its further development. The annotations and recommendations proposed in this work will allow for faster development and testing of classification software. As a result, this might facilitate cardiologists’ work and lead to faster diagnoses and earlier treatment.
ER  -

SMÍŠEK, R., Lucie MARŠÁNOVÁ, Andrea NĚMCOVÁ, Martin VÍTEK, Jiří KOZUMPLÍK and Marie NOVÁKOVÁ. CSE database: extended annotations and new recommendations for ECG software testing. \textit{Medical and Biological Engineering and Computing}. Heidelberg: Springer, 2017, vol.~55, No~8, p.~1473-1482. ISSN~0140-0118. Available from: https://dx.doi.org/10.1007/s11517-016-1607-5.

Detailed Information on Publication Record