pyhealth.tasks.readmission_prediction

class pyhealth.tasks.readmission_prediction.ReadmissionPredictionMIMIC3(window=datetime.timedelta(days=15), exclude_minors=True, **kwargs)

Bases: BaseTask

Readmission prediction on the MIMIC3 dataset.

This task aims at predicting whether the patient will be readmitted into hospital within a specified number of days based on clinical information from the current visit.

task_name

The name of the task.

Type:

str

input_schema

The schema for the task input.

Type:

Dict[str, str]

output_schema

The schema for the task output.

Type:

Dict[str, str]

Examples

>>> from pyhealth.datasets import MIMIC3Dataset
>>> from pyhealth.tasks import ReadmissionPredictionMIMIC3
>>> dataset = MIMIC3Dataset(
...     root="/path/to/mimic-iii/1.4",
...     tables=["diagnoses_icd", "procedures_icd", "prescriptions"],
... )
>>> task = ReadmissionPredictionMIMIC3()
>>> samples = dataset.set_task(task)

task_name: str = 'ReadmissionPredictionMIMIC3'

input_schema: Dict[str, str] = {'conditions': 'sequence', 'drugs': 'sequence', 'procedures': 'sequence'}

output_schema: Dict[str, str] = {'readmission': 'binary'}

pre_filter(df)

Return type:

LazyFrame

class pyhealth.tasks.readmission_prediction.ReadmissionPredictionMIMIC4(window=datetime.timedelta(days=15), exclude_minors=True, **kwargs)

Bases: BaseTask

Readmission prediction on the MIMIC4 dataset.

This task aims at predicting whether the patient will be readmitted into hospital within a specified number of days based on clinical information from the current visit.

task_name

The name of the task.

Type:

str

input_schema

The schema for the task input.

Type:

Dict[str, str]

output_schema

The schema for the task output.

Type:

Dict[str, str]

Examples

>>> from pyhealth.datasets import MIMIC4EHRDataset
>>> from pyhealth.tasks import ReadmissionPredictionMIMIC4
>>> dataset = MIMIC4EHRDataset(
...     root="/path/to/mimic-iv/2.2",
...     tables=["diagnoses_icd", "procedures_icd", "prescriptions"],
... )
>>> task = ReadmissionPredictionMIMIC4()
>>> samples = dataset.set_task(task)

task_name: str = 'ReadmissionPredictionMIMIC4'

input_schema: Dict[str, str] = {'conditions': 'sequence', 'drugs': 'sequence', 'procedures': 'sequence'}

output_schema: Dict[str, str] = {'readmission': 'binary'}

pre_filter(df)

Return type:

LazyFrame

class pyhealth.tasks.readmission_prediction.ReadmissionPredictionEICU(exclude_minors=True, **kwargs)

Bases: BaseTask

Readmission prediction on the eICU dataset.

This task aims at predicting whether the patient will be readmitted into the ICU during the same hospital stay based on clinical information from the current ICU visit.

Features: - using diagnosis table (ICD9CM and ICD10CM) as condition codes - using physicalexam table as procedure codes - using medication table as drugs codes

task_name

The name of the task.

Type:

str

input_schema

The schema for the task input.

Type:

Dict[str, str]

output_schema

The schema for the task output.

Type:

Dict[str, str]

exclude_minors

Whether to exclude patients whose age is less than 18.

Type:

bool

Examples

>>> from pyhealth.datasets import eICUDataset
>>> from pyhealth.tasks import ReadmissionPredictionEICU
>>> dataset = eICUDataset(
...     root="/path/to/eicu-crd/2.0",
...     tables=["diagnosis", "medication", "physicalexam"],
... )
>>> task = ReadmissionPredictionEICU()
>>> sample_dataset = dataset.set_task(task)

task_name: str = 'ReadmissionPredictionEICU'

input_schema: Dict[str, str] = {'conditions': 'sequence', 'drugs': 'sequence', 'procedures': 'sequence'}

output_schema: Dict[str, str] = {'readmission': 'binary'}

pre_filter(df)

Return type:

LazyFrame

class pyhealth.tasks.readmission_prediction.ReadmissionPredictionOMOP(window=datetime.timedelta(days=15), exclude_minors=True, **kwargs)

Bases: BaseTask

Readmission prediction on the OMOP dataset.

This task aims at predicting whether the patient will be readmitted into hospital within a specified number of days based on clinical information from the current visit.

task_name

The name of the task.

Type:

str

input_schema

The schema for the task input.

Type:

Dict[str, str]

output_schema

The schema for the task output.

Type:

Dict[str, str]

Examples

>>> from pyhealth.datasets import OMOPDataset
>>> from pyhealth.tasks import ReadmissionPredictionOMOP
>>> dataset = OMOPDataset(
...     root="/path/to/omop/data",
...     tables=["condition_occurrence", "procedure_occurrence",
...             "drug_exposure"],
... )
>>> task = ReadmissionPredictionOMOP()
>>> samples = dataset.set_task(task)

task_name: str = 'ReadmissionPredictionOMOP'

input_schema: Dict[str, str] = {'conditions': 'sequence', 'drugs': 'sequence', 'procedures': 'sequence'}

output_schema: Dict[str, str] = {'readmission': 'binary'}

pre_filter(df)

Return type:

LazyFrame