Comparing ChatGPT's and Surgeon's Responses to Thyroid-related Questions From Patients - Summary - MDSpire

Comparing ChatGPT's and Surgeon's Responses to Thyroid-related Questions From Patients

  • By

  • Siyin Guo

  • Ruicen Li

  • Genpeng Li

  • Wenjie Chen

  • Jing Huang

  • Linye He

  • Yu Ma

  • Liying Wang

  • Hongping Zheng

  • Chunxiang Tian

  • Yatong Zhao

  • Xinmin Pan

  • Hongxing Wan

  • Dasheng Liu

  • Zhihui Li

  • Jianyong Lei

  • April 10, 2024

  • 0 min

Share

Objective:

To assess the ability of ChatGPT (version GPT-4.0) to provide accurate, comprehensive, compassionate, and satisfactory responses to common thyroid-related questions, focusing on the quality of information and user experience.

Key Findings:
  • ChatGPT provided faster responses than both junior and senior specialists, with statistical significance (P < .001).
  • ChatGPT's responses were longer than those of both specialists, indicating a more detailed approach.
  • ChatGPT scored higher than both specialists in accuracy, comprehensiveness, compassion, and satisfaction, suggesting superior performance.
Interpretation:

ChatGPT outperformed human specialists in providing responses to common thyroid-related questions, suggesting its potential utility in enhancing patient education and support.

Limitations:
  • The study's sample size was limited to 30 questions, which may not represent all thyroid-related inquiries.
  • Further research is needed to validate ChatGPT's performance on complex thyroid questions and to address potential biases in evaluations.
Conclusion:

ChatGPT shows promise as a tool for addressing common thyroid-related inquiries, but further validation is necessary for more complex scenarios.

Original Source(s)

Related Content