Skip to main content

Table 1 Characteristics of the single-center and multicenter dataset

From: Contextual property detection in Dutch diagnosis descriptions for uncertainty, laterality and temporality

Dataset/characteristics Single-center dataset: Amsterdam UMC Multicenter dataset: five Dutch hospitals
Total records available, n 288,935 1,035,059
Modified descriptions, n(%) 73,280 (25.4) 175,210 (16.9)
Time period 1-1-2017–31-12-2017 28-4-2018–29-5-2019
Medical specialties, n 37 62 original; 41 after clustering
Usage for this study Development and internal validation of algorithm Multicenter validation of algorithm and to measure the frequency of types of contextual properties