Table 1 Illustrations of alternative types of data preprocessing filters applied to different types of clinical measurements
From: Modeling and prediction of pressure injury in hospitalized patients using artificial intelligence
Item | Date group/type | Data type/nature | Derived-data aggregation method |
|---|---|---|---|
1 | Biographical | Static/single | As is |
2 | Daily vital data | Time series (daily) | Min, max, mean |
3 | Braden (4 metrics) | Time series | min, max for each |
4 | Location/units | Unit names and dates | LOS for each location |
5 | Periop | Time series | avg, absolute min, absolute max |
6 | Surgical related data | Â | Â |
| Â | Â Anesthesia duration | Time series (minutely) | Total duration |
| Â | Â Procedure room stay | Time series (minutely) | Total duration |
| Â | Â Surgical schedule? | Categorical | As is? |
| Â | Â Surgical service? | Categorical | As is? |
7 | Staffing data | Time series | Mean (RN HPPD and Total HPPD) |
8 | ASA data (severity of illness) | Time series | Min, max, mean, SD |