https://creativecommons.org/publicdomain/zero/1.0/https://creativecommons.org/publicdomain/zero/1.0/
This dataset represents operational features, captured in four different hospital facilities, processing walk-in (F4) or scheduled (F1, F2, F3) patients. Approximately 600 to 1000 days of full patient flow data were extracted from each facility. The variables represent different features of the patient flow, captured at the time of patient arrivals and departure (one line per each patient visit event).
This data was extracted from the real hospital information system. Therefore it was anonymized and aggregated to remove any confidential information. In particular, x_ArrivalDTTM, x_ScheduledDTTM, and x_BeginDTTM timestamps were anonymized and set to the fictitious future dates (to make you aware of this modification). Only their relative timing was preserved. As a result, these timestamps can be used only for sorting, to reflect the correct order of feature events, but nothing else. This data is available for the operational experiment by a medical group of massachusetts general hospital to explore more.
Not seeing a result you expected?
Learn how you can add new datasets to our index.
https://creativecommons.org/publicdomain/zero/1.0/https://creativecommons.org/publicdomain/zero/1.0/
This dataset represents operational features, captured in four different hospital facilities, processing walk-in (F4) or scheduled (F1, F2, F3) patients. Approximately 600 to 1000 days of full patient flow data were extracted from each facility. The variables represent different features of the patient flow, captured at the time of patient arrivals and departure (one line per each patient visit event).
This data was extracted from the real hospital information system. Therefore it was anonymized and aggregated to remove any confidential information. In particular, x_ArrivalDTTM, x_ScheduledDTTM, and x_BeginDTTM timestamps were anonymized and set to the fictitious future dates (to make you aware of this modification). Only their relative timing was preserved. As a result, these timestamps can be used only for sorting, to reflect the correct order of feature events, but nothing else. This data is available for the operational experiment by a medical group of massachusetts general hospital to explore more.