LingxiDiag-16K: Synthetic Psychiatric Dialogue Dataset for Diagnostic AI

Approximately 16,000 synthetic electronic medical records and doctor-patient consultation dialogues designed for training and evaluating LLM-based psychiatric diagnostic decision support systems. The dataset was created by XuShihao6715 and its demographically aligned distributions are intended to reflect real-world clinical settings. The dataset page was last updated on February 10, 2026.

Use Cases

• Training diagnostic decision support systems based on synthetic psychiatric consultation dialogues.
• Evaluating model performance on clinical reasoning tasks using demographically aligned data.
• Benchmarking AI models for understanding and generating contextually appropriate psychiatric dialogue.

Strengths

• Contains approximately 16,000 synthetic records, providing a substantial corpus for model development.
• Designed with demographically aligned distributions to reflect real-world clinical settings.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• The synthetic nature of the data may not fully capture the complexity of real clinical interactions.

Provenance

Source

huggingface

Collection Method

Synthetically generated, likely for AI training purposes.

Freshness

Last updated 2026-02-10 07:51:31; freshness should be verified.