Readability Scores for AI-Generated Myofascial Pain Syndrome Content

A dataset containing readability, quality, and reliability scores for AI-generated medical content on Myofascial Pain Syndrome. The data was created by Yüksel Erkin and published on figshare in June 2026. It compares outputs from three AI models using six readability indices and four quality/reliability scales.

Use Cases

• Benchmarking AI model readability for patient-facing materials based on six different indices (FRES, FKGL, GFOG, CLI, ARI, SMOG).
• Comparing content quality and reliability across AI platforms using the GQS, EQIP, DISCERN, and JAMA scales mentioned in the study.
• Analyzing the correlation between linguistic complexity and medical reliability in AI-generated text as described in the research.
• Training models to assess or generate medical text that meets a target readability level, such as the 6th-grade standard referenced.

Strengths

• Scores derived from six established readability indices and four quality/reliability scales, providing multi-faceted evaluation.
• Content generated from 18 relevant keywords identified via Google Trends data, focusing on a specific medical condition.
• Data is licensed under CC-BY-4.0, permitting reuse with attribution.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is very small (9.5 KB), indicating limited scope.

Provenance

Source

Yüksel Erkin via figshare.

Collection Method

Responses generated by querying ChatGPT (GPT-5.2), Gemini 3 Flash, and Perplexity (Sonar-4 Large) with medical keywords; scores calculated by two independent observers.

Freshness

Last updated 2026-06-04 17:22:32; freshness should be verified.