Influence of the training set composition on the estimation performance of linear ECG-lead transformations.

Guldenring, Daniel; Finlay, Dewar D.; Bond, Raymond R.; Kennedy, Alan; Doggart, Peter; Janjua, Ghalib; McLaughlin, James

Abstract

Linear ECG-lead transformations (LELTs) are used to estimate unrecorded target leads by applying a number of recorded basis leads to a LELT matrix. Such LELT matrices are commonly developed using training datasets that are composed of ECGs that belong to different diagnostic classes (DCs). The aim of our research was to assess the influence of the training set composition on the estimation performance of LELTs that estimate target leads V1, V3, V4 and V6 from basis leads I, II, V2 and V5 of the 12-lead ECG. Our assessment was performed using ECGs from the three DCs left ventricular hypertrophy, right bundle branch block and normal (ECGs without abnormalities). Training sets with different DC compositions were used for the development of LELT matrices. These matrices were used to estimate the target leads of different test sets. The estimation performance of the developed matrices was quantified using root mean square error values calculated between derived and recorded target leads. Our findings indicate that unbalanced training sets can lead to LELTs that show large estimation performance variability across different DCs. Balanced training sets were found to produce LELTs that performed well across multiple DCs. We recommend balanced training sets for the development of LELTs.

Citation

GULDENRING, D., FINLAY, D.D., BOND, R.R., KENNEDY, A., DOGGART, P., JANJUA, G. and MCLAUGHLIN, J. 2023. Influence of the training set composition on the estimation performance of linear ECG-lead transformations. In Proceedings of the 50th Computing in cardiology 2023 (CinC 2023), 1-4 October 2023, Atlanta, GA, USA. Piscataway: IEEE/CinC [online], 50, article number 263. Available from: https://doi.org/10.22489/CinC.2023.263

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 50th Computing in cardiology 2023 (CinC 2023)
• Start Date: Oct 1, 2023
• End Date: Oct 4, 2023
• Acceptance Date: May 1, 2023
• Online Publication Date: Oct 3, 2023
• Publication Date: Dec 31, 2023
• Deposit Date: Jan 25, 2024
• Publicly Available Date: Jan 25, 2024
• Publisher: Institute of Electrical and Electronics Engineers (IEEE)
• Peer Reviewed: Peer Reviewed
• Series Title: Computing in cardiology
• Series Number: 50
• Series ISSN: 2325-8861
• ISBN: 9798350382525
• DOI: https://doi.org/10.22489/CinC.2023.263
• Keywords: Linear ECG-lead transformations (LELTs); Linear electrocardiographic (ECG)
• Public URL: https://rgu-repository.worktribe.com/output/2218670