ECGInstruct: Large-Scale Instruction-Tuning Dataset for ECG Images

ECGInstruct is a large-scale instruction-tuning dataset designed for electrocardiographic (ECG) images, created by the PULSE-ECG team. The dataset is associated with the research paper 'Teach Multimodal LLMs to Comprehend Electrocardiographic Images' and was last updated on the Hugging Face platform in October 2024.

Use Cases

• Instruction-tuning multimodal LLMs based on the described ECG image and instruction pairs.
• Benchmarking model performance on ECG comprehension tasks based on the associated ECGBench dataset.
• Research on aligning visual medical data with natural language instructions based on the dataset's stated purpose.

Strengths

• Dataset is explicitly designed as 'large-scale' for instruction-tuning.
• Directly linked to a peer-reviewed research paper (arXiv:2410.19008) and associated model (PULSE-7B).
• Last updated on 2024-10-28, indicating recent maintenance.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, file formats, and license information are unknown, which may limit suitability assessment.

Provenance

Source

PULSE-ECG team, associated with the AIMedLab.

Collection Method

Created for the research paper 'Teach Multimodal LLMs to Comprehend Electrocardiographic Images'; specific collection method not detailed.

Time Range

null

Freshness

Last updated 2024-10-28 01:30:35.

Geography

null