Table 1.
Patient Demographics and Distribution of Triage Categories in the Training and Validation Data Sets
| Training (n=336.835) | Validation (n=984) | |
|---|---|---|
| Male sex, n (%) | 188 858 (56.1) | 402 (54.3) |
| Age, mean (SD), y | 60.8 (15.5) | 60.4 (15.3) |
| Location, n (%) | ||
| Emergency department | 92 532 (27.5) | 310 (31.5) |
| Intensive care unit | 20 045 (6.0) | 63 (6.4) |
| Noncardiology outpatient clinic | 73 170 (21.7) | 161 (16.4) |
| Noncardiology ward | 86 630 (25.7) | 263 (26.7) |
| Preoperative screening | 6300 (1.9) | 8 (0.8) |
| Recovery ward | 53 994 (16.0) | 163 (16.6) |
| Other | 4164 (1.2) | 16 (1.6) |
| Triage category, n (%) | ||
| Normal | 142 456 (42.3)a | 418 (42.5)b |
| Abnormal, not acute | 153 360 (45.5)a | 410 (41.7)b |
| Abnormal, subacute | 24 731 (7.3)a | 76 (7.7)b |
| Abnormal, acute | 16 288 (4.8)a | 80 (8.1)b |
A 5% randomly sampled subset of the training data set was used for model tuning and internal validation. The validation data set is independent from the training data set.
a
Distribution based on text‐mining categorization of annotations by physician in daily practice.
b
Distribution based on the expert consensus panel annotations.