NidaanKosha 100K V1.0: Laboratory Investigation Data for 100,000 Indian Subjects

A dataset of laboratory investigations for 100,000 Indian subjects, containing over 6.8 million readings. The dataset, named NidaanKosha, is described as a treasury of diagnostic information. It was created by author 'ekacare' and last updated on the Hugging Face platform on 2025-03-16.

Use Cases

• Develop predictive models for disease diagnosis based on aggregated laboratory readings.
• Study population-level health trends and biomarker distributions within an Indian cohort.
• Train or benchmark clinical decision support systems using real-world diagnostic data.

Strengths

• Contains data from 100,000 subjects, providing a substantial sample size.
• Includes over 6.8 million individual laboratory readings, suggesting detailed per-subject records.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality requires manual inspection after download.
• The specific time range and geographic granularity within India are not provided.

Provenance

Source

huggingface

Collection Method

Likely collected from laboratory investigations, though the specific collection method is not detailed.

Freshness

Last updated 2025-03-16 03:56:29; freshness should be verified.

Geography

India