26 Clinical Tasks

Predict in-hospital mortality from ICU visit sequences of diagnoses and procedures.

from pyhealth.tasks import MortalityPredictionMIMIC3
samples = dataset.set_task(MortalityPredictionMIMIC3())

Binary mortality prediction on MIMIC-IV timeseries data using vitals and lab measurements.

from pyhealth.tasks import InHospitalMortalityMIMIC4
samples = dataset.set_task(InHospitalMortalityMIMIC4())

Predict whether a patient will be readmitted within 30 days of hospital discharge.

from pyhealth.tasks import ReadmissionPredictionMIMIC3
samples = dataset.set_task(ReadmissionPredictionMIMIC3())

Predict onset of diabetic ketoacidosis (DKA) from clinical event sequences in ICU patients.

from pyhealth.tasks import DKAPrediction
samples = dataset.set_task(DKAPrediction())

Classify clinical EEG recordings as normal or abnormal using deep signal models.

from pyhealth.tasks import EEGAbnormal
samples = dataset.set_task(EEGAbnormal())

Binary classification of chest X-rays as COVID-19 positive vs. normal/pneumonia.

from pyhealth.tasks import COVID19CXRBinary
samples = dataset.set_task(COVID19CXRBinary())

Classify genomic variants as pathogenic or benign using sequence-based features.

from pyhealth.tasks import MutationPathogenicity
samples = dataset.set_task(MutationPathogenicity())

Predict 5-year survival of prostate cancer patients from multi-omics profiles.

from pyhealth.tasks import CancerSurvivalPrediction
samples = dataset.set_task(CancerSurvivalPrediction())

Determine whether two clinical records belong to the same patient across fragmented health systems.

from pyhealth.tasks import PatientLinkage
samples = dataset.set_task(PatientLinkage())

Predict ICU length of stay as one of multiple duration buckets from admission data.

from pyhealth.tasks import LengthOfStayMIMIC3
samples = dataset.set_task(LengthOfStayMIMIC3())

Classify 30-second EEG epochs into Wake, N1, N2, N3, or REM sleep stages.

from pyhealth.tasks import SleepStagingSleepEDF
samples = dataset.set_task(SleepStagingSleepEDF())

5-class sleep staging on the large-scale Sleep Heart Health Study polysomnography dataset.

from pyhealth.tasks import SleepStagingSHHS
samples = dataset.set_task(SleepStagingSHHS())

Sleep stage classification on the ISRUC dataset including patients with sleep disorders.

from pyhealth.tasks import SleepStagingISRUC
samples = dataset.set_task(SleepStagingISRUC())

Classify EEG events into six types including spike-wave complexes, periodic discharges, and artifacts.

from pyhealth.tasks import EEGEvents
samples = dataset.set_task(EEGEvents())

3-class chest X-ray classification: COVID-19, viral pneumonia, or normal lung.

from pyhealth.tasks import COVID19CXRMulticlass
samples = dataset.set_task(COVID19CXRMulticlass())

Classify clinical transcription notes into one of 40 medical specialties using language models.

from pyhealth.tasks import MedicalTranscriptions
samples = dataset.set_task(MedicalTranscriptions())

15-task few-shot benchmark suite for evaluating clinical foundation models on real-world EHR prediction challenges.

from pyhealth.tasks import EHRShotTask
samples = dataset.set_task(EHRShotTask())

Recommend a safe set of medications for a patient visit given their diagnosis and procedure history.

from pyhealth.tasks import DrugRecommendationMIMIC3
samples = dataset.set_task(DrugRecommendationMIMIC3())

Automatically assign ICD-9 diagnosis codes from clinical notes and structured EHR data.

from pyhealth.tasks import ICD9CodingMIMIC3
samples = dataset.set_task(ICD9CodingMIMIC3())

Detect multiple cardiac arrhythmia types simultaneously from 12-lead ECG recordings.

from pyhealth.tasks import CardioDetection
samples = dataset.set_task(CardioDetection())

Classify cardiac valve disease conditions from phonocardiogram (PCG) heart sound recordings.

from pyhealth.tasks import HeartSoundClassification
samples = dataset.set_task(HeartSoundClassification())

Multi-label classification of 14 thoracic diseases from frontal chest X-ray images.

from pyhealth.tasks import ChestXray14Multilabel
samples = dataset.set_task(ChestXray14Multilabel())

Predict functional impact and cancer driver gene status of somatic mutations.

from pyhealth.tasks import VariantClassification
samples = dataset.set_task(VariantClassification())

Recommend medication combinations that minimize drug-drug interaction (DDI) risk for a patient visit.

from pyhealth.tasks import DrugRecommendationMIMIC3
samples = dataset.set_task(DrugRecommendationMIMIC3())

Medication recommendation from multi-center ICU data with cross-site generalization evaluation.

from pyhealth.tasks import DrugRecommendationeICU
samples = dataset.set_task(DrugRecommendationeICU())

Predict tumor mutation burden (TMB) as a continuous value from multi-omics cancer profiles.

from pyhealth.tasks import CancerMutationBurden
samples = dataset.set_task(CancerMutationBurden())