Datasets:

NLP-FBK
/

dyspnea-clinical-notes

dataset_info:
  features:
    - name: clinical_note
      dtype: string
    - name: document_id
      dtype: string
  splits:
    - name: en
      num_bytes: 2673782
      num_examples: 2667
    - name: it
      num_bytes: 2663860
      num_examples: 2667
  download_size: 3045357
  dataset_size: 5337642
configs:
  - config_name: default
    data_files:
      - split: en
        path: data/en-*
      - split: it
        path: data/it-*

Dataset description

This dataset contains the train unannotated clinical notes for the CRF:filling Shared Task at CL4Health2026. The clinical notes have been collected, anonymized and annotated at the San Giovanni Bosco (SGB) hospital, Turin, Italy. There are two splits, each representing a different language: en (English) and it (Italian). English data has been automatically translated from Italian.

Each example (2667 in total) in the dataset is composed by:

document_id: clinical note identifier
clinical_note: the note reporting on the patient's clinical history

For more info visit the website https://sites.google.com/fbk.eu/crf