LungCURE: 1,000 Real-World Clinical Cases for Lung Cancer AI Benchmarking

LungCURE is the first standardized multimodal benchmark for evaluating AI models on precision lung cancer clinical decision support. It comprises 1,000 real-world, doctor-labeled clinical cases collected across more than ten hospitals in China. The dataset was created by author Fine2378 and last updated on Hugging Face in April 2026.

Use Cases

• Benchmarking multimodal AI models on clinical reasoning tasks based on real-world patient cases.
• Training AI assistants for lung cancer diagnosis and treatment planning based on multimodal clinical data.
• Researching the application of large language models in precision oncology based on doctor-labeled case data.

Strengths

• Contains 1,000 real-world clinical cases.
• Data is doctor-labeled, suggesting expert validation.
• Cases were collected from more than ten hospitals, indicating multi-institutional scope.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Data may reflect geographic bias inherent to its collection from hospitals in China.

Provenance

Source

More than ten hospitals in China.

Collection Method

Constructed from real-world, doctor-labeled clinical cases.

Freshness

Last updated 2026-04-09 07:47:13; freshness should be verified.

Geography

China