The integration of artificial intelligence (AI) tools in various sectors, including ophthalmology, has introduced innovative solutions. In the specialized field of ophthalmology, effective communication and patient education pose significant challenges, especially for conditions like glaucoma. The advent of AI chatbots, such as ChatGPT, offers a promising avenue for enhancing patient education in glaucoma care.

Unlocking the Potential of AI Chatbots

ChatGPT, developed by OpenAI, exemplifies the capabilities of AI chatbots. It simulates human-like conversations, providing accurate and credible responses to medical inquiries. Notably, ChatGPT has shown proficiency in responding to queries from glaucoma specialists, as highlighted by Saif Aldeen AlRyalat, MD, and others.

ChatGPT’s Performance in Ophthalmic Triage

A study by Tsui et al. assessed ChatGPT Feb 13 version’s ability to triage ophthalmic symptoms. Out of ten common ocular conditions tested, ChatGPT provided precise and suitable responses in eight instances. However, it faltered in scenarios requiring urgency or detailed inquiries, underscoring the need for further advancements in AI chatbots for medical triage.

Advancements with GPT-4

A recent study has re-evaluated the effectiveness of GPT-4, an advanced version of ChatGPT, in triaging these same ophthalmic symptoms. Waisberg et al. found that GPT-4 improved upon the limitations of its predecessor, offering precise and suitable responses to complex medical scenarios, including macular degeneration and other eye-related issues.

Leveraging ChatGPT for Patient Engagement

The efficacy of ChatGPT in patient education was explored in a study involving glaucoma-related questions, with the Flesch Kincaid Grade Level Score being used to assess comprehension. The chatbot’s responses were found to be easily understandable for individuals with a high school education or above.

Quality of AI Chatbot-Generated Advice

A study compared the quality of ophthalmology advice from AI chatbots to human physicians. Expert reviewers discerned responses generated by humans versus chatbots with an accuracy rate of 61%, with both sets showing similar levels of correctness and alignment with medical consensus.

Navigating AI Chatbots in Glaucoma Care

AI chatbots can distill complex medical jargon into understandable information, bridging the gap between specialized language and patient understanding. However, they should be complementary tools, not replacements for human healthcare providers.

Envisioning the Future with AI Chatbots

AI chatbots hold transformative potential in reshaping patient education within glaucoma care and beyond. However, their successful integration hinges on ethical considerations and upholding the human touch in healthcare.

A demonstration “Glaucoma Education Chatbot” using ChatGPT can be found here

References:

Ophthalmology.” Glaucoma Physician, December 2022.

Jonathan C. Tsui, et al. “Appropriateness of ophthalmic symptoms triage by a popular online artificial intelligence chatbot.” Eye, 2023.

Ethan Waisberg, et al. “GPT-4 for triaging ophthalmic symptoms.” Eye, 2023.

David A Lee, et al. “Can ChatGPT™, an ‘intelligent’ chatbot, be used to educate our glaucoma patients?” Invest. Ophthalmol. Vis. Sci., 2023.

Bernstein IA, et al. “Comparison of Ophthalmologist and Large Language Model Chatbot Responses to Online Patient Eye Care Questions.” JAMA Netw Open, 2023.