pyhealth.data.Patient#

Another basic data structure in the package. A Patient is a collection of Visit for the current patients. It contains all necessary attributes of a patient, such as ethnicity, mortality status, gender, etc. It can support various healthcare tasks.

class pyhealth.data.Patient(patient_id, data_source)[source]#

Bases: object

Patient class representing a sequence of events.

patient_id#

Unique patient identifier.

Type:

str

data_source#

DataFrame containing all events, sorted by timestamp.

Type:

pl.DataFrame

event_type_partitions#

Dictionary mapping event types to their respective DataFrame partitions.

Type:

Dict[str, pl.DataFrame]

get_events(event_type=None, start=None, end=None, filters=None, return_df=False)[source]#

Get events with optional type and time filters.

Parameters:

  • event_type (Optional[str]) – Type of events to filter.

  • start (Optional[datetime]) – Start time for filtering events.

  • end (Optional[datetime]) – End time for filtering events.

  • return_df (bool) – Whether to return a DataFrame or a list of Event objects.

  • filters (Optional[List[tuple]]) – Additional filters as [(attr, op, value), …], e.g.: [(“attr1”, “!=”, “abnormal”), (“attr2”, “!=”, 1)]. Filters are applied after type and time filters. The logic is “AND” between different filters.

Returns:

Filtered events as a DataFrame or a list of Event objects.

Return type:

Union[pl.DataFrame, List[Event]]