AI Dialogue Comparisons for Cartilage Repair Questions

A study comparing ChatGPT, DeepSeek, and Google Search on 20 cartilage repair questions across tissue engineering and surgery domains. The dataset includes blinded quality scoring using the Accuracy-Safety-Hallucination framework and readability analysis via the Flesch-Kincaid formula. Authored by Sen Yang Xiao and last updated on 2026-05-29 under a CC-BY-4.0 license.

Use Cases

• Benchmarking AI model accuracy and safety in medical domains based on the described ASH framework scoring.
• Analyzing the readability of medical information from different sources based on Flesch-Kincaid Grade Level scores.
• Comparing the functional roles of different AI tools for specific clinical stakeholders as suggested by the study's conclusions.

Strengths

• Direct comparison of three platforms (ChatGPT, DeepSeek, Google) on identical questions.
• Independent evaluation by three blinded raters using a structured framework.
• Results include specific median accuracy scores and statistical significance (p-values).

Limitations

• Dataset is a 22.9 KB DOCX file, indicating a very limited scope, likely containing summary results rather than raw data.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

figshare

Collection Method

Questions were sourced from top Google Search FAQs in cartilage tissue engineering (2023) and cartilage repair surgery (2024), then submitted to all three platforms for matched comparison.

Time Range

Questions sourced from 2023 and 2024.

Freshness

Last updated 2026-05-29 05:53:27; freshness should be verified.