Atrial fibrillation classification using step-by-step machine learning

被引：14

作者：

Goodfellow, Sebastian D. ^{[1
]}

Goodwin, Andrew ^{[1
]}

Greer, Robert ^{[1
]}

Laussen, Peter C. ^{[1
]}

Mazwi, Mjaye ^{[1
]}

Eytan, Danny ^{[1
,2
]}

机构：

[1] Hosp Sick Children, Toronto, ON, Canada

[2] Rambam Med Ctr, Haifa, Israel

来源：

BIOMEDICAL PHYSICS & ENGINEERING EXPRESS | 2018年 / 4卷 / 04期

关键词：

machine learning; signal processing; ECG Waveforms; physionet challenge;

D O I：

10.1088/2057-1976/aabef4

中图分类号：

R8 [特种医学]; R445 [影像诊断学];

学科分类号：

1002 ; 100207 ; 1009 ;

摘要：

This paper presents a detailed overview of our submission to the 2017 Physionet Challenge where competitors were asked to build a model to classify a single lead ECG waveform as either normal sinus rhythm, atrial fibrillation, other rhythm, or noisy. A step-by-step machine learning pipeline was assembled, which included signal conditioning, R-peak detection and filtering, and feature extraction. Asuite of over 300 features, falling into one of three main feature groups; template features, RRI features, and full waveform features, were extracted from each waveform and an XGBoost, tree-based, gradient boosting classifier was used as the machine learning algorithm. The model produced a cross-validation F-1 score of 0.8245, a hidden sub-test score of 0.82, and a hidden test score of 0.8125. The score breakdown for each class (normal sinus rhythm, atrial fibrillation, other rhythm, and noisy) was as follows: F-1,F- NRS = 0.9024, F-1,F- AF = 0.8156, F-1,F- OR = 0.7194, F-1,F- Noise =. 0.5705.

引用

页数：16

共 27 条

[1] Optimal parameters study for sample entropy-based atrial fibrillation organization analysis [J].