Will a robot handle your office visits? Not likely, but AI can improve patient care and reduce costs
Artificial intelligence is working its way into every aspect of our lives, and so it is natural that the technology will move into the field of eye care as well.
Currently, artificial intelligence (AI) is being used in fields as diverse as agriculture and farming, flying, security and surveillance, self-driving cars, sports analytics, retail and fashion, and warehousing and logistics supply chains. When used correctly, AI can reduce human error, take on risks so humans don’t have to, offer around-the-clock support, make decisions faster, uncover efficiencies, and save money.
In healthcare, AI is playing increasingly important and visible roles, and if it is not yet having an impact on your eye care practice, just wait a minute — chances are, it soon will. In healthcare settings, artificial intelligence is helping improve diagnostic accuracy, analyze health care data, issue clinical alerts, efficiently manage health care services, and even reconstruct patient history. Application areas are clustered around five main regions of the industry: health services management, predictive medicine, patient data, diagnostics, and clinical decision-making. Combined, this is helping health care providers deliver more accurate and informed care faster and at lower costs.
AI’s purpose “is not to eliminate the ophthalmologist, but to enhance and increase access for patients,” said Robert Chang, MD, an expert in the field. Eye care practices shouldn’t fear the technology, but rather, they should embrace it as a way to make their practices more efficient and effective. As the repetitive tasks that providers perform regularly are eliminated, those working in health care will be able to focus more on doing the things that AI can’t do—like providing personal care to patients and building relationships.
What is artificial intelligence? A primer for those in health care
What is artificial intelligence? Pop culture often portrays AI as a robot that appears suspiciously human. While some AI applications are in robotics, robots are unlikely to perform cataract surgery anytime soon. At its core, AI is simply an algorithm — it is a set of rules that a computer follows to solve a problem or perform tasks that would normally require human intelligence. In addition to applications in industry, AI is becoming more common in everyday life, and not just for the wealthy. Applications you may be personally familiar with include:
- Voice recognition (Alexa)
- Personalization of your social media feeds
- Options presented to you on Netflix, Spotify, or YouTube
- Online ads that seem to read your mind.
- Navigation maps like Waze or Google Maps
- Customer service chatbots that handle issues as well as a human does
It is common to see AI appear alongside a similar term — machine learning. The two are sometimes used interchangeably, but there are key differences. Machine learning is the tool that powers artificial intelligence; it involves the creation of algorithms that can learn on their own by using extremely large data sets and eventually apply its observations to new data sets to then make decisions or predictions on its own.
AI is currently being used successfully to diagnose diabetic retinopathy. An algorithm is fed large data sets consisting of fundus or OCT photos where diabetic retinopathy is present. Eventually, the algorithm learns what disease characteristics to look for, and can diagnose the disease in new photos. Outside of eye care, is being applied to improve lung cancer diagnostics, including the detection of lung cancer.
Four ways AI is transforming eye care
Will AI force you out of work? Not anytime soon. Will it have a direct impact on your practice? Probably. Here are four ways AI is transforming eye care right now:
- AI is helping in diagnostics. AI-driven tools can offer a higher level of objectivity and precision when analyzing medical images.
- AI is becoming a medical partner. New AI tools, like those from Google’s DeepMind, don’t just offer a diagnosis, they offer an explanation behind the diagnosis.
- AI is helping health care providers save time. AI is handling repetitive work, freeing up more time for health care providers, and even performing more complicated tasks such as creating three-dimensional models of ophthalmologic tumors by marking up data from two-dimensional image scans.
- AI is creating new ways for patients to seek care. AI-powered hardware and software solutions are able to detect diseases such as diabetic retinopathy and glaucoma, and are proving to be more accurate at such diagnoses than trained ophthalmologists are.
Will artificial intelligence become your screening partner?
AI is revolutionizing eye care on a number of fronts. It is becoming increasingly important as populations age globally, leading to growing numbers of people who live with vision impairment, blindness, and major eye diseases. It’s also helping to expand the physical abilities of eye clinics and hospitals and extend the expert availability of eye care providers such as optometrist and ophthalmologists, a trend which is occurring both in developing and developed nations.
Screening for diabetic retinopathy and diabetic macular edema is the first area to benefit from advances in AI. AI and its ophthalmological applications are extending and expanding the use of digital ophthalmology and improving the accessibility, availability, and productivity of existing resources. Paired with telemedicine, AI is decentralizing and scaling eye screening and primary care. AI is working its way into eye screening thanks to well-understood relationships between clinical features and disease severity in major eye disease such as glaucoma.
The classification of ophthalmic imaging data such as color fundus photography using DL algorithms has been described in the literature, as has the classification of age-related macular degeneration (AMD) in color fundus photography to predict AMD progression using optical coherence tomography (OCT) and to classify glaucoma in imaging.
Similarly, if primary care doctors, optometrists, or endocrinologists could screen for DR in their offices, they could screen a higher percentage of patients with diabetes and the patients they did send to ophthalmology practices would be far more likely to actually have DR. Recently, the Food and Drug Administration approved technology that can do just that — the IDx-DR, which relies on fundus photos. The technology employs deep learning in collaboration with retina specialists to potentially bring the retina specialist’s diagnostic expertise directly into primary care settings.
Will AI increase accessibility and help more patients receive care?
Along with telemedicine, AI screenings are seen as ways for rural residents to more easily receive care — and that means not just rural areas in America but in also in developing nations in places like Africa and South America. AI-assisted DR screening is being used to address the acute shortage of specialists in rural areas, helping to expand care while reducing competition among physicians in urban areas. Some 90 percent of patients with diabetes in rural areas aren’t getting DR screenings.
Artificial intelligence: Caveats to be aware of
Yes, AI is cool and powerful. No, it can’t do everything. Here are some caveats to consider when thinking about AI and eye care:
- An AI algorithm only looks for what you tell it to look for and might miss things that a human would notice.
- If a variable changes, the algorithm will have to re-learn.
- Researchers sometimes can’t explain why one algorithm works when another doesn’t.
For eye care, AI is an exciting new frontier, and it’s being put to use on more than the likes of AMD – currently, AI is also being used for retinopathy of prematurity (ROP), macular degeneration, glaucoma, and identifying refractive error from fundus photos. What is next? It will be exciting to see where this field goes.
