DentalGemma Instruct: 2,494 Synthetic Dental Clinical Cases for AI Fine-Tuning

2,494 synthetic text-only clinical cases formatted for instruction tuning, designed for fine-tuning the MedGemma 1.5 4B IT model on dental diagnosis and treatment planning. The dataset covers 98 unique dental conditions and is split into 2,246 training and 248 validation samples. It was created by naazimsnh02 and last updated on February 23, 2026.

Use Cases

• Fine-tuning language models for dental diagnosis based on synthetic clinical case descriptions.
• Instruction-tuning models for dental treatment planning based on the covered conditions.
• Benchmarking medical AI models on a specialized dental text dataset.
• Training conversational AI assistants for dental clinical education or simulation.

Strengths

• 2,494 total samples provide a substantial base for fine-tuning.
• Covers 98 unique dental conditions, suggesting diversity in clinical scenarios.
• Explicitly designed for a specific model (MedGemma 1.5 4B IT), indicating a clear use case.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Data is synthetic, which may limit realism and generalizability to real-world clinical data.

Provenance

Source

Derived from the Wildstash/dental-2.5k-instruct dataset.

Collection Method

Synthetically generated for instruction tuning.

Time Range

null

Freshness

Last updated 2026-02-23 00:07:55; freshness should be verified.

Geography

null