Standardized Database of 12-Lead Electrocardiograms with a Common Standard for the Promotion of Cardiovascular Research: KURIAS-ECG

• DOI: 10.4258/hir.2023.29.2.132
• Publication Date: 2023-05-09T00:08:00Z
• Authors: Hakje Yoo; Yunjin Yum; Soo Wan Park; Jeong Moon Lee; Moonjoung Jang; Yoojoong Kim; Jong-Ho Kim; Hyun-Joon Park; Kap Su Han; Jae Hyoung Park; Hyung Joon Joo
• Keywords: Lead (geology); Promotion (chess)

Objectives: Electrocardiography (ECG)-based diagnosis by experts cannot maintain uniform quality because individual differences may occur. Previous public databases can be used for clinical studies, but there is no common standard that would allow to combined. For this reason, it difficult conduct research derives results combining databases. Recent commercial ECG machines offer diagnoses similar those of a physician. Therefore, the purpose study was construct standardized database using computerized diagnoses.Methods: The constructed Systematized Nomenclature Medicine Clinical Terms (SNOMED CT) and Observational Medical Outcomes Partnership–common data model (OMOP-CDM), were then categorized into 10 groups based on Minnesota classification. In addition, extract high-quality waveforms, poor-quality ECGs removed, bias minimized extracting at least 2,000 cases each group. To check quality, difference in baseline displacement according whether poor removed analyzed, usefulness verified with seven classification models waveforms.Results: KURIAS-ECG consists from 13,862 patients, about 20,000 points, making possible obtain more than An artificial intelligence extracted through SNOMED-CT showed an average accuracy 88.03%.Conclusions: contains various machines. proposed protocol should promote cardiovascular disease big intelligence.