Learning From Alarms : A Robust Learning Approach for Accurate Photoplethysmography-Based Atrial Fibrillation Detection using Eight Million Samples Labeled with Imprecise Arrhythmia Alarms
Atrial fibrillation (AF) is a common cardiac arrhythmia with serious health consequences if not detected and treated early. Detecting AF using wearable devices with photoplethysmography (PPG) sensors and deep neural networks has demonstrated some success using proprietary algorithms in commercial solutions. However, to improve continuous AF detection in ambulatory settings towards a population-wide screening use case, we face several challenges, one of which is the lack of large-scale labeled training data. To address this challenge, we propose to leverage AF alarms from bedside patient monitors to label concurrent PPG signals, resulting in the largest PPG-AF dataset so far (8.5M 30-second records from 24,100 patients) and demonstrating a practical approach to build large labeled PPG datasets. Furthermore, we recognize that the AF labels thus obtained contain errors because of false AF alarms generated from imperfect built-in algorithms from bedside monitors. Dealing with label noise with unknown distribution characteristics in this case requires advanced algorithms. We, therefore, introduce and open-source a novel loss design, the cluster membership consistency (CMC) loss, to mitigate label errors. By comparing CMC with state-of-the-art methods selected from a noisy label competition, we demonstrate its superiority in handling label noise in PPG data, resilience to poor-quality signals, and computational efficiency.
Medienart: |
E-Artikel |
---|
Erscheinungsjahr: |
2024 |
---|---|
Erschienen: |
2024 |
Enthalten in: |
Zur Gesamtaufnahme - volume:PP |
---|---|
Enthalten in: |
IEEE journal of biomedical and health informatics - PP(2024) vom: 01. Feb. |
Sprache: |
Englisch |
---|
Beteiligte Personen: |
Ding, Cheng [VerfasserIn] |
---|
Links: |
---|
Themen: |
---|
Anmerkungen: |
Date Revised 01.02.2024 published: Print-Electronic Citation Status Publisher |
---|
doi: |
10.1109/JBHI.2024.3360952 |
---|
funding: |
|
---|---|
Förderinstitution / Projekttitel: |
|
PPN (Katalog-ID): |
NLM367901765 |
---|
LEADER | 01000naa a22002652 4500 | ||
---|---|---|---|
001 | NLM367901765 | ||
003 | DE-627 | ||
005 | 20240202232254.0 | ||
007 | cr uuu---uuuuu | ||
008 | 240202s2024 xx |||||o 00| ||eng c | ||
024 | 7 | |a 10.1109/JBHI.2024.3360952 |2 doi | |
028 | 5 | 2 | |a pubmed24n1278.xml |
035 | |a (DE-627)NLM367901765 | ||
035 | |a (NLM)38300786 | ||
040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
041 | |a eng | ||
100 | 1 | |a Ding, Cheng |e verfasserin |4 aut | |
245 | 1 | 0 | |a Learning From Alarms |b A Robust Learning Approach for Accurate Photoplethysmography-Based Atrial Fibrillation Detection using Eight Million Samples Labeled with Imprecise Arrhythmia Alarms |
264 | 1 | |c 2024 | |
336 | |a Text |b txt |2 rdacontent | ||
337 | |a ƒaComputermedien |b c |2 rdamedia | ||
338 | |a ƒa Online-Ressource |b cr |2 rdacarrier | ||
500 | |a Date Revised 01.02.2024 | ||
500 | |a published: Print-Electronic | ||
500 | |a Citation Status Publisher | ||
520 | |a Atrial fibrillation (AF) is a common cardiac arrhythmia with serious health consequences if not detected and treated early. Detecting AF using wearable devices with photoplethysmography (PPG) sensors and deep neural networks has demonstrated some success using proprietary algorithms in commercial solutions. However, to improve continuous AF detection in ambulatory settings towards a population-wide screening use case, we face several challenges, one of which is the lack of large-scale labeled training data. To address this challenge, we propose to leverage AF alarms from bedside patient monitors to label concurrent PPG signals, resulting in the largest PPG-AF dataset so far (8.5M 30-second records from 24,100 patients) and demonstrating a practical approach to build large labeled PPG datasets. Furthermore, we recognize that the AF labels thus obtained contain errors because of false AF alarms generated from imperfect built-in algorithms from bedside monitors. Dealing with label noise with unknown distribution characteristics in this case requires advanced algorithms. We, therefore, introduce and open-source a novel loss design, the cluster membership consistency (CMC) loss, to mitigate label errors. By comparing CMC with state-of-the-art methods selected from a noisy label competition, we demonstrate its superiority in handling label noise in PPG data, resilience to poor-quality signals, and computational efficiency | ||
650 | 4 | |a Journal Article | |
700 | 1 | |a Guo, Zhicheng |e verfasserin |4 aut | |
700 | 1 | |a Rudin, Cynthia |e verfasserin |4 aut | |
700 | 1 | |a Xiao, Ran |e verfasserin |4 aut | |
700 | 1 | |a Shah, Amit |e verfasserin |4 aut | |
700 | 1 | |a Do, Duc H |e verfasserin |4 aut | |
700 | 1 | |a Lee, Randall J |e verfasserin |4 aut | |
700 | 1 | |a Clifford, Gari |e verfasserin |4 aut | |
700 | 1 | |a Nahab, Fadi B |e verfasserin |4 aut | |
700 | 1 | |a Hu, Xiao |e verfasserin |4 aut | |
773 | 0 | 8 | |i Enthalten in |t IEEE journal of biomedical and health informatics |d 2013 |g PP(2024) vom: 01. Feb. |w (DE-627)NLM217081614 |x 2168-2208 |7 nnns |
773 | 1 | 8 | |g volume:PP |g year:2024 |g day:01 |g month:02 |
856 | 4 | 0 | |u http://dx.doi.org/10.1109/JBHI.2024.3360952 |3 Volltext |
912 | |a GBV_USEFLAG_A | ||
912 | |a GBV_NLM | ||
951 | |a AR | ||
952 | |d PP |j 2024 |b 01 |c 02 |