DermoInstruct: Large-Scale Dermatology Visual Instruction Tuning Dataset

DermoInstruct is a large-scale dermatology-oriented visual instruction tuning dataset for multimodal medical AI research. It contains image-grounded instruction-response conversations designed to support tasks such as lesion description, morphology recognition, and diagnostic reasoning. The dataset was created by author mendicant04 and was last updated on 2026-05-14.

Use Cases

• Fine-tuning vision-language models for dermatology based on image-grounded conversations.
• Training models for lesion description and morphology recognition based on visual instruction data.
• Developing systems for diagnostic reasoning and clinical question answering in dermatology.

Strengths

• Designed as a large-scale dataset specifically for dermatology and dermoscopy tasks.
• Structured in a LLaVA/Qwen-VL-style conversation format intended for supervised fine-tuning.

Limitations

• Row count, file formats, and column-level documentation are unknown, limiting suitability assessment.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

huggingface

Freshness

Last updated 2026-05-14 12:59:11; freshness should be verified.