ICD-10 Clinical Notes: Multilingual Diagnosis Codes and Sample Notes

34 languages provide official ICD-10 diagnosis names in this dataset from Uppsala University. It includes sample clinical journal notes in English and Swedish and is structured with train/test splits for classifier training. The dataset was last updated on Hugging Face in January 2026.

Use Cases

• Train a multilingual ICD-10 code classifier based on the official diagnosis names in 34 languages.
• Develop clinical note generation or understanding models based on the provided English and Swedish sample notes.
• Benchmark cross-lingual transfer learning for medical NLP using the provided language coverage.
• Create educational or reference tools for medical coders using the standardized diagnosis terminology.

Strengths

• Covers 34 languages, including 24 EU and 10 major world languages.
• Includes structured train/test splits intended for classifier training.
• Provides sample clinical journal notes, suggesting practical application data.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Birger Moëll, Department of Linguistics and Philology, Uppsala University

Collection Method

Likely compiled from official ICD-10 resources and clinical sources.

Time Range

null

Freshness

Last updated 2026-01-23 14:49:23; freshness should be verified.

Geography

null