Artificial intelligence (AI) is growing fast in healthcare, and eye care is one of the fields it is affecting. In the United States, many eye care workers like ophthalmologists, optometrists, and clinic managers are thinking about how to use AI tools to help patients. AI can help with diagnosing problems, planning treatments, and managing patients. This brings both good chances and some challenges. For healthcare leaders, owners, and IT workers, knowing about these changes is important for better eye care while keeping tasks organized and following rules.
This article looks at recent studies on how AI has worked in managing eye patients, shows some worries and needs of doctors and patients, and talks about adding AI into eye care routines.
Recent research shows AI systems, especially those using big language models and deep learning, can work as well as or better than human eye specialists. A study from the New York Eye and Ear Infirmary of Mount Sinai, published in JAMA Ophthalmology, compared GPT-4, a smart AI chatbot, to expert glaucoma and retina doctors.
The study gave 20 tough eye-care case questions about glaucoma and retina problems to both AI and human experts. Their answers were rated for accuracy and how complete they were. GPT-4 scored higher than glaucoma specialists and about the same as retina experts. It also gave more complete answers than the glaucoma doctors and was close to the retina doctors. This shows AI can help support doctors by giving good diagnoses and treatment ideas.
Dr. Louis R. Pasquale, who led the study, said AI did better than expected with tough cases. He thinks GPT-4 can help doctors get better at diagnosing and planning treatments. Dr. Andy Huang, the lead author, said AI might help more people get good eye care faster by giving expert advice when needed.
The research suggests smart AI tools could help improve treatment for eye diseases like glaucoma and retinal problems by giving clear, evidence-based guidance.
Even with good results, using AI widely in eye care still faces some problems. The Asia-Pacific Academy of Ophthalmology notes that many AI programs have not yet gotten official approval for use in real eye clinics. This is because there is not enough proof they work well, it is hard to see how they make decisions, and it’s tricky to fit AI into current medical routines.
Eye care workers worry about how reliable AI answers are and the risks of depending too much on machines. They also think about legal duties and privacy of patient data. AI needs to work well with systems already used in clinics, like electronic health records and imaging tools.
A good way to make AI work is to involve doctors and staff early when building and using these tools. This helps AI meet real needs, fit into daily work smoothly, and keep doctors in control of key decisions.
Patients can also be unsure about using AI in their care. They worry about their privacy and want to understand how AI makes suggestions. To get patients to accept AI, clear and honest information about its benefits and limits is needed.
Clinics must balance new technology with ethics, such as protecting patient privacy and making sure AI decisions keep care quality high.
Besides helping with diagnosis, AI is being used to automate office work and clinic tasks. This is important especially in busy US eye clinics to handle many patients and cut down on extra work.
Some offices use AI to answer phones and set appointments. For example, companies like Simbo AI offer AI phone services that handle questions, book visits, and give instructions. This frees up staff and makes work run smoother.
This helps clinic managers and IT teams who want solutions that reduce patient wait times and lower frustration from slow phone systems.
AI helps not just office jobs but also clinical tasks. EyeArt, an AI system made by Eyenuk, can screen for diabetic eye disease quickly during visits. It looks at retinal images and gives results in a minute without needing a specialist or pupil dilation. This lets doctors or trained staff check many patients faster.
EyeArt has been approved by the US FDA and in Europe. Data from over 100,000 patients shows it detects diabetic eye problems correctly over 91% of the time, and serious cases 98.5% of the time. This means AI can safely help doctors find issues early and prevent vision loss.
AI lets nurses and technicians handle screening and data tasks independently, freeing eye doctors to focus on harder cases. This helps clinics see more patients and reduce wait times, which is important as more people need care.
Good AI use also supports telemedicine and remote patient checks, which have grown since COVID-19, giving more options to serve patients.
AI models that use both pictures and text, like GPT-4 with visual features (GPT-4 V), show promise for future eye care by mixing image study and patient history. Researchers at Chaim Sheba Medical Center studied how GPT-4 V did diagnosing eye conditions using photos and clinical info.
GPT-4 V was right about 47.5% of the time when it saw only images. Adding patient info like age and symptoms helped accuracy go up to 67.5%. This was close to two doctors who were not eye specialists, who scored between 60% and 72.5% with that info.
The study was small and did not use some advanced testing, but it showed that AI can think like humans by using both pictures and text together.
The authors said multimodal AI could help with decision-making, choosing research groups, and teaching doctors. But issues like patient privacy, data safety, and bias need attention before these tools can be widely used.
For US eye clinic leaders and owners, knowing about AI is important to make smart choices about buying tech and planning operations.
Using AI to automate tasks helps clinics run better beyond just diagnosis. It improves administration and patient communication.
Services like those from Simbo AI use AI to answer phones, book visits, confirm appointments, and answer questions without staff. This cuts down on staff work and shortens wait times on calls, which matters in busy practices.
This automation also stops lost income from missed or wrong appointments by keeping booking accurate. Staff can then focus on harder tasks needing a human touch.
AI that works with clinic software helps patients check in by sending reminders and collecting info. It can also book follow-ups or tests based on AI diagnosis suggestions, like for diabetic retinopathy.
AI helps doctors by writing notes and assisting with paperwork. This saves time and helps doctors keep better records, improving care and billing.
For IT teams, it is key to make sure AI works smoothly with electronic health records and imaging systems to keep workflows steady and data safe.
Using AI in eye care shows clear improvements in diagnosis and patient management, especially for glaucoma and retinal diseases. Studies show AI tools like GPT-4 can do as well or better than human experts in tough cases. Systems like EyeArt work fast and alone to screen for diabetic eye disease, and they have official approval for real use in the US.
Still, some problems need solving to use AI in daily practice, like acceptance by doctors and patients, official approvals, privacy concerns, and fitting with current technology.
For US eye clinic leaders, careful planning is needed to use AI well while fixing these issues. Office automation tools like Simbo AI’s phone services can help run clinics better alongside clinical AI tools. Together, these can improve patient flow, staff use, and eye care quality.
As AI grows, ongoing checks, staff training, and patient involvement will be important to build lasting AI-supported eye care in the United States.
The study found that AI, specifically the GPT-4 chatbot, was able to match or outperform human specialists in the management of glaucoma and retinal disease based on diagnostic accuracy and comprehensiveness.
Researchers presented ophthalmological case management questions to the GPT-4 chatbot and compared its responses with those of fellowship-trained glaucoma and retina specialists, scoring them on a Likert scale for accuracy and completeness.
The mean rank for accuracy was 506.2 for the LLM chatbot vs. 403.4 for glaucoma specialists, and for completeness it was 528.3 vs. 398.7, showing significant improvements in the AI’s performance.
The Dunn test showed significant differences in ratings between the AI and specialist performance for diagnostic accuracy and completeness, except in the case of specialist vs. trainee ratings.
The study suggests that AI could play a significant role in diagnosing and managing glaucoma and retinal diseases, potentially serving as a supportive tool for eyecare providers.
AI tools like GPT-4 can provide guidance on documentation and clinical decision-making, helping ophthalmologists improve their clinical practices in patient management.
Dr. Pasquale highlighted that AI’s proficiency in handling patient cases was surprising and that it could enhance clinician skills, similar to how Grammarly aids writers.
Dr. Huang noted that the performance of GPT-4 was eye-opening and indicated the massive potential for AI systems in enhancing clinical practices for seasoned specialists.
The lead author acknowledged that while the findings are promising, additional testing is needed to validate the AI’s performance before it can be fully integrated into clinical settings.
Integration of AI could lead to faster access to expert advice for patients, resulting in more informed decision-making and potentially improved treatment outcomes.
Simbo is using AI agents to automate all phone related workflows. Connect now to know more.
Schedule AI Agents Demo Now© Since 2019, Simbo AI.
