Review Reports
by
- Carlos Ruiz-Núñez 1,†,
- Javier Gismero Rodríguez 2,† and
- Iván Herrera-Peco 6,7,8,*,†
- et al.
Reviewer 1: Takanobu Hirosawa Reviewer 2: Susana Henriques
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsGeneral comments
=============
Thank you for submitting your manuscript, "Can Generative AI Contribute to Health Literacy? A Study in the Field of Ophthalmology." The topic is both intriguing and timely, reflecting significant potential in the integration of generative AI within ophthalmology. However, the manuscript requires substantial revisions in terms of structure and clarity to effectively communicate its findings and relevance to the fields of digital health and AI in medicine.
Specific comments
=============
Major comments
---------------------
- Overall:
- Please update all references to ChatGPT-3.5, as the service has been discontinued. Current references should reflect the most recent versions such as ChatGPT-4.
- Ensure consistent terminology when referring to different versions of the ChatGPT model to avoid confusion.
- I recommend adopting a checklist similar to the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) to improve the manuscript’s structure and comprehensiveness.
- Introduction:
- Enhance the focus on the significance of ophthalmology literacy and the role of generative AI.
- Clearly identify the existing gaps this study addresses and articulate the study's aims, particularly emphasizing the need to compare results between expert and non-expert users.
- Methods:
- Detail the selection process for the panel of ophthalmologists and the development and validation of the 12 initial questions.
- Clarify the operational definitions of "non-expert user" and "expert user," particularly in terms of zero-shot versus few-shot learning paradigms.
- Explain the expansion of the study from 12 questions to 42 questionnaires and outline the primary outcomes of this expansion.
- Results:
- Include representative outputs from the generative AI model used, ensuring that these examples are relevant and illustrative of the points being made.
- Discussion:
- Discuss the significant findings in detail, addressing how they contribute to the field of digital health and AI.
- Acknowledge the limitations related to the scale of the study, particularly the choice of only 12 initial questions and their impact on the study's generalizability.
- Conclusion:
- Clarify conclusions drawn regarding the statistical significance of the findings, particularly in terms of the reliability and usefulness of the AI-generated responses as evaluated by expert consultations.
Addressing these points will significantly enhance the manuscript's quality, relevance, and potential impact on the field.
Comments on the Quality of English LanguagePlease refer my previous comments and the rate in the Quality of English Language.
Author Response
Dear Reviewer,
Thank you very much for your suggestions and feedback, which have improved the article. Please find attached our responses to your comments. The changes made to the manuscript are highlighted in red to make them easier to identify.
Kind regards.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsDear authors,
The topic is udated and very relevant, and the article is well organised.
The authors start from a premise (presented right in the introduction) which, across the board, jeopardises the work. The idea that a tool like ChatGTP is a condition for improving the health literacy (in this case) of users. It's not!
Health literacy is a complex process, of which access to quality and relevant information is only the foundation. It also involves: understanding this information; assessing the importance and appropriateness of the information; and mobilising and applying the relevant information to a specific situation. However, the study presented does not go beyond the first condition by demonstrating that ChatGTP is able to produce reliable information regarding the main questions of ophthalmological health patients. If it doesn't go beyond providing access to quality information, it cannot be concluded that ChatGTP contributes to improving the health literacy of these users.
This is a limitation that must be overcome for the manuscript to be considered for publication, in accordance with the journal's quality standards. Therefore, either the quality of the available information is assumed to be improved or it is reformulated to show an effective increase in user literacy.
Kind Regards
Author Response
Dear Reviewer,
Thank you very much for your suggestions and feedback, which have improved the article. Please find attached our responses to your comments. The changes made to the manuscript are highlighted in red to make them easier to identify.
Kind regards.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsGeneral comments
=============
Thank you for submitting your manuscript, "Can Generative AI Contribute to Health Literacy? A Study in the Field of Ophthalmology." The topic is both intriguing and timely, reflecting significant potential in the integration of generative AI within ophthalmology. Almost all responses were reasonable.