Sign in Subscribe
Cancer research

Artificial Intelligence Tools Show Potential in Disseminating Medical Information on Prostate Cancer

Strategic Insights

— 1 min read

Research conducted at Capital Medical University has evaluated the quality, accuracy, and readability of prostate-cancer-related medical information produced by ChatGPT and DeepSeek. The study found that while both AI tools demonstrated potential for disseminating medical information, there may be variability in the quality and readability of the content generated. The research aimed to assess the understandability and actionability of AI-generated content, as well as evaluate the quality of treatment-related information. The study collected frequently asked questions related to prostate cancer from the American Cancer Society website, ChatGPT, and DeepSeek, and had three urologists review and confirm the relevance of the selected questions.

Key Takeaways:

  • The study aimed to evaluate the quality, accuracy, and readability of prostate-cancer-related medical information produced by ChatGPT and DeepSeek.
  • The Patient Education Materials Assessment Tool for Printable Materials (PEMAT-P) was used to assess the understandability and actionability of AI-generated content, with ChatGPT and DeepSeek receiving scores of 70.66±8.13 and 69.35±8.83, respectively.
  • The DISCERN instrument was used to evaluate the quality of treatment-related information, with both ChatGPT and DeepSeek receiving scores of 59.07±3.39 and 58.88±3.66, respectively.
  • The Automated Readability Index (ARI) for DeepSeek was higher than that for ChatGPT, indicating a potential difference in readability.
  • The study concluded that further refinement of content quality and language clarity is needed to prevent potential misunderstandings, decisional uncertainty, and anxiety among patients due to difficulty in comprehension.
  • The research involved collecting frequently asked questions related to prostate cancer from the American Cancer Society website, ChatGPT, and DeepSeek, and having three urologists review and confirm the relevance of the selected questions.
  • The study collected data from a comprehensive evaluation of AI-generated content, including the use of established indices to assess readability.

Statistics:

  • The PEMAT-P scores for ChatGPT and DeepSeek were 70.66±8.13 and 69.35±8.83, respectively.
  • The DISCERN scores for ChatGPT and DeepSeek were 59.07±3.39 and 58.88±3.66, respectively.
  • The ARI for DeepSeek was higher than that for ChatGPT, at 12.63±1.42 and 10.85±1.93, respectively.

Sources:

  • NewsRx LLC
  • Cancer Weekly
  • The World Journal of Men's Health
  • Capital Medical University
  • Beijing Anzhen Nanchong Hospital
  • Korean Soc Sexual Medicine & Andrology
  • Pusan Natl Univ Medical Sch
  • Dept Urology
  • 179 Gudeok-Ro, Seo-Gu, Busan, South Korea

Read more