Validating an SVM-based neonatal seizure detection algorithm for generalizability, non-inferiority and clinical efficacy

Show full item record



Permalink

http://hdl.handle.net/10138/346299

Citation

Tapani , K T , Nevalainen , P , Vanhatalo , S & Stevenson , N J 2022 , ' Validating an SVM-based neonatal seizure detection algorithm for generalizability, non-inferiority and clinical efficacy ' , Computers in Biology and Medicine , vol. 145 , 105399 . https://doi.org/10.1016/j.compbiomed.2022.105399

Title: Validating an SVM-based neonatal seizure detection algorithm for generalizability, non-inferiority and clinical efficacy
Author: Tapani, Karoliina T.; Nevalainen, Päivi; Vanhatalo, Sampsa; Stevenson, Nathan J.
Contributor organization: HUSLAB
University of Helsinki
HUS Children and Adolescents
Children's Hospital
Kymsote – Social and Health Services in Kymenlaakso
HYKS erva
Clinicum
HUS Medical Imaging Center
BioMag Laboratory
Department of Neurosciences
HUS Diagnostic Center
Kliinisen neurofysiologian yksikkö
Department of Physiology
Date: 2022-06
Language: eng
Number of pages: 10
Belongs to series: Computers in Biology and Medicine
ISSN: 0010-4825
DOI: https://doi.org/10.1016/j.compbiomed.2022.105399
URI: http://hdl.handle.net/10138/346299
Abstract: Neonatal seizure detection algorithms (SDA) are approaching the benchmark of human expert annotation. Measures of algorithm generalizability and non-inferiority as well as measures of clinical efficacy are needed to assess the full scope of neonatal SDA performance. We validated our neonatal SDA on an independent data set of 28 neonates. Generalizability was tested by comparing the performance of the original training set (cross -validation) to its performance on the validation set. Non-inferiority was tested by assessing inter-observer agreement between combinations of SDA and two human expert annotations. Clinical efficacy was tested by comparing how the SDA and human experts quantified seizure burden and identified clinically significant periods of seizure activity in the EEG. Algorithm performance was consistent between training and validation sets with no significant worsening in AUC (p > 0.05, n = 28). SDA output was inferior to the annotation of the human expert, however, re-training with an increased diversity of data resulted in non-inferior performance (delta kappa = 0.077, 95% CI:-0.002-0.232, n = 18). The SDA assessment of seizure burden had an accuracy ranging from 89 to 93%, and 87% for identifying periods of clinical interest. The proposed SDA is approaching human equivalence and provides a clinically relevant interpretation of the EEG.
Subject: neonatal EEG
EEG monitoring
Neonatal intensive care unit
Seizure
Support vector machine
AGREEMENT
3126 Surgery, anesthesiology, intensive care, radiology
Peer reviewed: Yes
Rights: cc_by
Usage restriction: openAccess
Self-archived version: publishedVersion


Files in this item

Total number of downloads: Loading...

Files Size Format View
1_s2.0_S0010482522001913_main.pdf 4.937Mb PDF View/Open

This item appears in the following Collection(s)

Show full item record