Detection of atrial fibrillation using decision tree ensemble

Guanguo Bin; Minggang Shao; Guanghong Bin; Jiao Huang; Dingchang Zheng; Shuicai Wu

doi:10.22489/CinC.2017.342-204

Detection of atrial fibrillation using decision tree ensemble

Guanguo Bin, Minggang Shao, Guanghong Bin, Jiao Huang, Dingchang Zheng, Shuicai Wu

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding › peer-review

38 Citations (Scopus)

Abstract

2017 PhysioNet/CinC Challenge proposed a global competition for classifying a short single ECG lead recording into normal sinus rhythm, atrial fibrillation (AF), alternative rhythm, and unclassified rhythm. This study developed and evaluated a pragmatic approach to solve the challenge, which was based on a decision tree ensemble with 30 features from ECG recording. The model was trained using the AdaBoost.M2 algorithm. The results reported here were obtained using 100-fold cross-validation, and the lowest MSE was 0.12 with the maximum number of splits of 55, and the number of trees of 20. The entry was tested and scored in the second phase of the challenge. The achieved scores for "Normal", "AF", "Other", were 0.93, 0.86, and 0.79, respectively, while the F1 measure was 0.86, and the official overall score was 0.82.

Original language	English
Title of host publication	Computing in Cardiology
Publisher	IEEE
ISBN (Electronic)	978-1-5386-6630-2
ISBN (Print)	978-1-5386-4555-0
DOIs	https://doi.org/10.22489/CinC.2017.342-204
Publication status	Published - 5 Apr 2018
Externally published	Yes
Event	Computing in Cardiology 2017 - Rennes, France Duration: 24 Sept 2017 → 27 Sept 2017 Conference number: 44 https://dblp.org/db/conf/cinc/cinc2017

Conference

Conference	Computing in Cardiology 2017
Abbreviated title	CINC 2017
Country/Territory	France
City	Rennes
Period	24/09/17 → 27/09/17
Internet address	https://dblp.org/db/conf/cinc/cinc2017

Access to Document

10.22489/CinC.2017.342-204

Cite this

@inproceedings{57ec55454ef24fb9bb098fcb057dd9f6,

title = "Detection of atrial fibrillation using decision tree ensemble",

abstract = "2017 PhysioNet/CinC Challenge proposed a global competition for classifying a short single ECG lead recording into normal sinus rhythm, atrial fibrillation (AF), alternative rhythm, and unclassified rhythm. This study developed and evaluated a pragmatic approach to solve the challenge, which was based on a decision tree ensemble with 30 features from ECG recording. The model was trained using the AdaBoost.M2 algorithm. The results reported here were obtained using 100-fold cross-validation, and the lowest MSE was 0.12 with the maximum number of splits of 55, and the number of trees of 20. The entry was tested and scored in the second phase of the challenge. The achieved scores for {"}Normal{"}, {"}AF{"}, {"}Other{"}, were 0.93, 0.86, and 0.79, respectively, while the F1 measure was 0.86, and the official overall score was 0.82.",

author = "Guanguo Bin and Minggang Shao and Guanghong Bin and Jiao Huang and Dingchang Zheng and Shuicai Wu",

year = "2018",

month = apr,

day = "5",

doi = "10.22489/CinC.2017.342-204",

language = "English",

isbn = "978-1-5386-4555-0",

booktitle = "Computing in Cardiology",

publisher = "IEEE",

address = "United States",

note = "Computing in Cardiology 2017, CINC 2017 ; Conference date: 24-09-2017 Through 27-09-2017",

url = "https://dblp.org/db/conf/cinc/cinc2017",

}

TY - GEN

T1 - Detection of atrial fibrillation using decision tree ensemble

AU - Bin, Guanguo

AU - Shao, Minggang

AU - Bin, Guanghong

AU - Huang, Jiao

AU - Zheng, Dingchang

AU - Wu, Shuicai

N1 - Conference code: 44

PY - 2018/4/5

Y1 - 2018/4/5

N2 - 2017 PhysioNet/CinC Challenge proposed a global competition for classifying a short single ECG lead recording into normal sinus rhythm, atrial fibrillation (AF), alternative rhythm, and unclassified rhythm. This study developed and evaluated a pragmatic approach to solve the challenge, which was based on a decision tree ensemble with 30 features from ECG recording. The model was trained using the AdaBoost.M2 algorithm. The results reported here were obtained using 100-fold cross-validation, and the lowest MSE was 0.12 with the maximum number of splits of 55, and the number of trees of 20. The entry was tested and scored in the second phase of the challenge. The achieved scores for "Normal", "AF", "Other", were 0.93, 0.86, and 0.79, respectively, while the F1 measure was 0.86, and the official overall score was 0.82.

AB - 2017 PhysioNet/CinC Challenge proposed a global competition for classifying a short single ECG lead recording into normal sinus rhythm, atrial fibrillation (AF), alternative rhythm, and unclassified rhythm. This study developed and evaluated a pragmatic approach to solve the challenge, which was based on a decision tree ensemble with 30 features from ECG recording. The model was trained using the AdaBoost.M2 algorithm. The results reported here were obtained using 100-fold cross-validation, and the lowest MSE was 0.12 with the maximum number of splits of 55, and the number of trees of 20. The entry was tested and scored in the second phase of the challenge. The achieved scores for "Normal", "AF", "Other", were 0.93, 0.86, and 0.79, respectively, while the F1 measure was 0.86, and the official overall score was 0.82.

UR - http://www.scopus.com/inward/record.url?eid=2-s2.0-85045091480&partnerID=MN8TOARS

U2 - 10.22489/CinC.2017.342-204

DO - 10.22489/CinC.2017.342-204

M3 - Conference proceeding

SN - 978-1-5386-4555-0

BT - Computing in Cardiology

PB - IEEE

T2 - Computing in Cardiology 2017

Y2 - 24 September 2017 through 27 September 2017

ER -

Detection of atrial fibrillation using decision tree ensemble

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this