Detection of Short Supraventricular Tachycardias in Single-lead ECGs Recorded Using a Handheld Device
(2022) 2022 Computing in Cardiology, CinC 2022- Abstract
Short supraventricular tachycardias (S-SVTs) have been associated with a higher risk of developing atrial fibrillation (AF). Hence, identification of participants with such arrhythmias may increase the yield of AF screening. However, the lower signal quality of ECGs recorded using handheld screening devices challenges the detection of S-SVT. In the present work, a new method for detection of S-SVT is presented, which is based on the requirement on morphologic similarity between the detected beats. Specifically, any episode with a sequence of beats of similar morphology is considered as an S-SVT candidate while any episode with detections of different morphology, either due to signal disturbances or aberrant ectopic beats, is excluded.... (More)
Short supraventricular tachycardias (S-SVTs) have been associated with a higher risk of developing atrial fibrillation (AF). Hence, identification of participants with such arrhythmias may increase the yield of AF screening. However, the lower signal quality of ECGs recorded using handheld screening devices challenges the detection of S-SVT. In the present work, a new method for detection of S-SVT is presented, which is based on the requirement on morphologic similarity between the detected beats. Specifically, any episode with a sequence of beats of similar morphology is considered as an S-SVT candidate while any episode with detections of different morphology, either due to signal disturbances or aberrant ectopic beats, is excluded. For this purpose, a support vector machine (SVM) was trained and validated, using a simulated ECG database, to classify an episode as either consisting of beats of similar or non-similar morphologies. Episodes identified as S-SVT candidates are subject to two further rhythm criteria in order to confirm the presence of an S-SVT. The performance of the S-SVT detector is evaluated using a subset of the StrokeStop I database (305 S-SVT out of 8258), resulting in a sensitivity, specificity, and positive predictive value of 88.8%, 92.0%, and 29.9%, respectively. In conclusion, the results suggest that the detection of S-SVT in AF screening can be done at an acceptable balance between sensitivity and positive predictive value.
(Less)
- author
- Halvaei, Hesam LU ; Hygrell, Tove ; Svennberg, Emma ; Corino, Valentina D.A. ; Sornmo, Leif LU and Stridh, Martin LU
- organization
- publishing date
- 2022
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- host publication
- Computing in Cardiology, CinC 2022
- publisher
- IEEE Computer Society
- conference name
- 2022 Computing in Cardiology, CinC 2022
- conference location
- Tampere, Finland
- conference dates
- 2022-09-04 - 2022-09-07
- external identifiers
-
- scopus:85152956994
- ISBN
- 9798350300970
- DOI
- 10.22489/CinC.2022.314
- language
- English
- LU publication?
- yes
- id
- 8f63bd13-3cbc-4560-b768-0c9df59524a4
- date added to LUP
- 2023-08-15 10:45:01
- date last changed
- 2024-02-19 22:31:13
@inproceedings{8f63bd13-3cbc-4560-b768-0c9df59524a4, abstract = {{<p>Short supraventricular tachycardias (S-SVTs) have been associated with a higher risk of developing atrial fibrillation (AF). Hence, identification of participants with such arrhythmias may increase the yield of AF screening. However, the lower signal quality of ECGs recorded using handheld screening devices challenges the detection of S-SVT. In the present work, a new method for detection of S-SVT is presented, which is based on the requirement on morphologic similarity between the detected beats. Specifically, any episode with a sequence of beats of similar morphology is considered as an S-SVT candidate while any episode with detections of different morphology, either due to signal disturbances or aberrant ectopic beats, is excluded. For this purpose, a support vector machine (SVM) was trained and validated, using a simulated ECG database, to classify an episode as either consisting of beats of similar or non-similar morphologies. Episodes identified as S-SVT candidates are subject to two further rhythm criteria in order to confirm the presence of an S-SVT. The performance of the S-SVT detector is evaluated using a subset of the StrokeStop I database (305 S-SVT out of 8258), resulting in a sensitivity, specificity, and positive predictive value of 88.8%, 92.0%, and 29.9%, respectively. In conclusion, the results suggest that the detection of S-SVT in AF screening can be done at an acceptable balance between sensitivity and positive predictive value.</p>}}, author = {{Halvaei, Hesam and Hygrell, Tove and Svennberg, Emma and Corino, Valentina D.A. and Sornmo, Leif and Stridh, Martin}}, booktitle = {{Computing in Cardiology, CinC 2022}}, isbn = {{9798350300970}}, language = {{eng}}, publisher = {{IEEE Computer Society}}, title = {{Detection of Short Supraventricular Tachycardias in Single-lead ECGs Recorded Using a Handheld Device}}, url = {{http://dx.doi.org/10.22489/CinC.2022.314}}, doi = {{10.22489/CinC.2022.314}}, year = {{2022}}, }