Table 2 Introduction of public datasets

Dataset Name

Dataset range

Major use

Capnobase [57]

42 cases of 8-min recordings, including capnograms and PPG

patient classification [38]

Meteor [58]

over 2 million unique patients with demographics, pathologies, imaging, laboratory tests, physical examinations, comorbidities, and treatment regimens, as well as financial and administrative information details

prediction of readmission risk [44]

Clinical Practice Research Datalink [59]

60 million patients with demographic characteristics, diagnoses and symptoms, drug exposures, vaccination history, laboratory tests and referrals to hospital and specialist care

symptoms prediction [47]

Dataset Name

Dataset range

Major use

Freesound Dataset Kaggle 2018 [60]

11,073 audio files

Prediction [50]

Respiratory Sound Database [61]

920 annotated audio samples from 126 subjects

classification of breath sounds [55]