IN 2023, there has been an increase in stories about artificial intelligence (AI). Whether watching the news or scrolling through social media, someone is talking about ChatGPT or how to use AI to simplify tasks.
The reality of AI in our world is accelerating. Many see it as much bigger in some capacities than it really is, while others minimize its daily impact. As eyecare practitioners, we have already seen AI make an impact in our universe—but maybe not in the way that you think.
As a futurist, I have been observing AI’s evolution and explosion in the last few years (and it is challenging to keep up). But I realize that many eyecare practitioners still do not fully comprehend what the term “artificial intelligence” really means. It is important to lay a foundation to understand the current platforms of AI and look down the runway to see where it might take us. Lastly, let’s gaze into the future and consider how practitioners may be using AI in eye care and, more specifically, the contact lens space in the coming year.
As a frequent columnist for Contact Lens Spectrum, this has been one of the most challenging articles to compose. AI is all over the place, but there are few places to find traditional research pointing toward its application in eye care, let alone the contact lens space. That doesn’t mean that it isn’t being used or investigated. Instead, it is showing up, like AI does, in organic ways.
Unlike other technologies that have launched quickly, AI has been evolving and inching its way into daily life and health care for nearly three-quarters of a century. Although it is part of our everyday life, rarely can we point out all the ways it impacts us. This evolution will continue right in front of us at a dynamic rate.
WHAT IS ARTIFICIAL INTELLIGENCE?
While AI is exploding, it is not a new science. Its origins date back to the 1950s and 1960s, when innovators like Alan Turing proposed that machines could develop intelligence. And over the last 70 years, we have seen innovation continue and grow to get us to where we are today.
AI is a complex field that has many disciplines. These subfields work to make intelligent systems (or processes) and machines that simplify things. Some of the more prominent subfields of AI include machine learning, deep learning, computer vision, robotics, expert systems, speech recognition, planning and scheduling, neural networks, cognitive computing, and natural language processing. Some of these may sound familiar, while others seem rather abstract.
Machine learning may be the most recognizable. It has existed for years in systems that we use daily. With machine learning, computers learn patterns from inputs and can make predictions or choices (think online streaming platforms helping you choose what to watch next).
A subset of machine learning that rises to the next level of using multiple layers of data is called deep learning. Deep learning uses a very complex network that has many layers—taking machine learning and other subsets of AI to their extreme. You can think of a machine taking thousands of data points from a topographer (computer vision), designing a scleral lens (expert systems), feeding it into a lathe, and then producing and shipping a lens to your office (robotics).
The language model is most likely the aspect of AI that has captured the attention of people today because of ChatGPT. ChatGPT is a product of OpenAI and the GPT stands for Generative Pre-trained Transformer. According to the company, its mission is to ensure that artificial general intelligence (AGI), which refers to highly autonomous systems that can outperform humans at most economically valuable work, benefits all of humanity.
OpenAI also offers an application programming interface (API) that other companies and developers can use. This allows them to integrate the complex language models into their own applications or services that have more advanced AI functions.
Many other mainstream tech companies are investing heavily in AI. And certainly, one can’t leave out research institutes and the advancements that they are making in the AI world. Of course, this data is always changing and new startup companies and research institutes are forming all the time and innovating at a rocket’s pace.
AI AND HEALTH CARE ETHICS
AI has a huge place in health care, and it’s important that innovators constantly look for methods and research on how to apply it for patient care. There are many ethical considerations when applying AI to health care. Ethical conversations cannot become stagnant when it comes to implementing AI into patients’ lives. Two very important considerations pertaining to AI and any health care practice include:
- The health of patients must stay at the forefront of innovation. Health cannot be sacrificed for innovation.
- The functionality of providers and patients must be enhanced so that health care can be more accessible. Innovation in AI must propel the field forward for the advancement of providers and for every patient.
AI AND CONTACT LENSES
Contact lenses have been evolving over the last century in incredible ways. A trip through the Contact Lens Museum in Forest Grove, OR, reveals the fast progression of our contact lens industry (thecontactlensmuseum.org ). And over the last few decades, there have been rumblings about innovations that are not yet available, but will take contact lenses to extreme heights beyond vision correction alone. These technologies include smart contact lenses, health monitoring, disease-detecting lenses, and augmented reality lenses.
AI also has the potential to devise more personalized fitting approaches to contact lenses and advances in personalized vision correction lenses that can take the clarity of vision to a new level. All these advances may further develop our standard spherical, toric, and multifocal lenses and even revolutionize specialty GP, scleral, and myopia management lens designs.
• Smart Contact Lenses: Merging Technology and Visual Optics In a generic sense, smart contact lenses utilize contact lenses as a base and then add innovations that reach beyond vision. These innovations extend into health monitoring and disease detection and treatment to visual enhancements beyond the realm of standard correction.
Smart lenses are different from standard lenses in that they contain components not found in standard lenses. These may include biomaterials, microfluidics, biosensors, power supply, data transmission, and display circuits.1
• Electronics in Contact Lenses When thinking about electronics in contact lenses, the advancements in engineering that are occurring to make this possible are incredible. While the specifications are beyond the scope of this article, it is important to know that some companies are developing contact lenses that have a battery supply (can hold power), wireless transmission (can send data or collect data), and a display.1,2
While we are seemingly still years away from having these devices fully operational in daily life, the innovations that they will make, combined with AI, is mind-blowing in its potential. While “smart contact lenses” that do not use electronics are being investigated and researched, the greatest innovation will be having electronics incorporated into contact lenses and AI adding its flair to maximize their benefits.
• Drug Delivery For decades, eyecare practitioners have been anticipating the release of drug delivery contact lenses. Contact lenses seem like the holy grail for drug delivery, especially for complex conditions like glaucoma, allergy, and dry eye. When a patient needs a long-standing medication, it only makes sense that delivering the drug to them in small doses over a long period of time will better help them stabilize the condition.
Fortunately, this dream is starting to become a reality with the release and approval of a contact lens that delivers ketotifen for allergies. While the concept is not new, many companies have tried—or are attempting to—get drug delivery CLs to market, but they are facing obstacles.
While it may be a long time before coming to fruition, a future that includes CL drug delivery will be maximized when AI plays a role in its application. With electronic biosensing and health monitoring, the contact lenses will be given a signal to release the exact dosage of medication when it is needed.
• Biosensing and Health Monitoring Tears are complex in the human body. They contain many properties that reveal what is occurring in the rest of the body. Contact lens devices are being designed to measure and take advantage of the tear properties and eye structure. The idea of using contact lenses for the measurement of lactic acid, glucose, and intraocular pressure (IOP) is currently being examined. This expansion will revolutionize how lenses are used for health monitoring and benefits.
IOP is measured every day using a standard method of applanation, but work continues to look for better ways of measuring IOP with the use of contact lenses.3 As such, researchers have been developing contact lenses that incorporate smart technology so they can track a patient’s IOP.4,5
For obvious reasons, frequent IOP measurements would give clinicians precise knowledge of what is occurring with their patient and could help them better handle and measure their patient’s progress with treatment. When these devices become more mainstream, they could be combined with drug delivery and AI in an all-in-one device that could measure and track a patient’s glaucoma, give data points to the patient and the clinicians, modify a treatment for a patient (wherever they are on the planet), and deliver medication to the patient.
Glucose monitoring has incredible medical capabilities, most notably for the diabetic patient base. Similarities have been shown in measuring glucose with the finger prick versus the tear film.6 This has led several companies to research and develop contact lenses that can measure and calculate patient glucose levels.
While many can conceptualize ways that this capability could be used, AI can be incorporated into the data that is collected. Given advanced calculations of a patient’s diet and activities over the span of weeks and months, an AI system could dispense medication to accurately and precisely control patients’ glucose levels and their diabetes.
Additionally, as seen on social media channels, glucose monitors are being marketed and sold to consumers to help them better understand their glucose spikes for diet and health benefits. If glucose-monitoring contact lenses become more mainstream, AI may be capable of guiding consumers toward proper diets and eating times that will help them reach their health goals more quickly and accurately.
• Visual Displays Most of us are aware of virtual reality (VR) and its ability to take us to totally new worlds where we can explore through video games, learn through three-dimensional visualization, or interact for health advancements (vision therapy). But a lesser known or utilized aspect of virtual reality is augmented reality (AR).
A mainstream application that became popular was Pokémon Go, which was released in 2016. We could hold our phones up and see the real world that we were living in with a false (or augmented) world placed on top of it. While this was one of the first mainstream uses of AR to become popular, this technology has very limited application because you have to hold up a device in order to make your AR world appear.
Knowing that handheld devices had limitations, one major tech company released a spectacle-based technology in 2013. While the introduction of the technology was a promising step in AR innovation, it ultimately failed. Critics believe its failure was because of its price ($1,500), battery life (the device only lasted about four hours), and its lack of functionality. Ultimately, it was pulled in January 2015 before major strides were made available for its release.7
The impact of AI-guided AR contact lenses on daily life will be the largest impact that contact lenses will ever make, in my opinion. With the integration of AI, displays will no longer be necessary for work; imagine just staring at a wall that displays a computer screen or attending a baseball game and seeing the statistics of the players as they run around the field. There will no longer be a need to look down at a screen for GPS in your car; it will be right in front of you highlighting the road to take. Unfortunately, as we have observed with major technology, advertisements and promotions may play a large role on our lives as they become more specifically targeted, but the possibilities of this technology are endless.
However, the advent of AR contact lenses still seems slightly out of reach, at least in the near term. Innovation and technology still have considerable steps to take, especially because of government regulation and the measures of safety that the health care sector needs to investigate. As Nathan Efron, AC, DSc, PhD, states: “Translating biomedical research into clinical reality can be a long and drawn-out process, interrupted by numerous obstacles and setbacks arising from the actions of a multitude of stakeholders such as companies, researchers, funders, policymakers, and regulators.”8
AI’S IMPACT ON SPECIALTY LENSES
Research is still building in the arena of AI’s impact on specialty lenses. However, there are obvious applications that we can hypothesize will be or are already in development.
AI machine learning capability is excellent at communicating with robotics, and this has great application for topography and lathes. We can predict that corneal topography and scleral profilometry will one day capture images for patients and then communicate with lathes at our laboratories to customize the lenses that we order for patients. The technology should be able to monitor the fit of the lenses through photography analysis, measuring refractive outcomes (through autorefractors) or analyzing amount of clearance, fluorescein pattern, and refractive data to learn how to modify the lens fits over time with optical coherence tomography imaging.
Additionally, future tear chemistry evaluations may be able to monitor the eye’s response to the contact lens and suggest modifications to solution or lens material to maximize the ocular surface outcome. Of course, there are many other ways that AI can maximize the benefits of modifications, and it will be exciting to see these unfold.
AI’S IMPACT ON MYOPIA MANAGEMENT
Similar to what has been discussed above, the fitting of orthokeratology lenses will be fine-tuned through advanced AI technology. For advancement of myopia management, data about the patient’s axial length, peripheral refractive data, pupil size, and age could be used to modify the soft multifocal, orthokeratology, concentration of atropine, or spectacle lens selection.
AI’S IMPACT ON STANDARD REFRACTIVE CONTACT LENSES
There are some very good contact lens treatments for our patients. In some cases, contact lenses can provide a better visual outcome for patients than the use of their glasses, while other patients may find that their toric or multifocal lenses yield outcomes that are slightly inferior to their spectacle lenses.
With devices connected to our slit lamps measuring a patient’s topography and tear chemistry, advanced refractive devices that measure aberrations, AI will be able to more precisely recommend the most ideal contact lens prescription, lens material, cleaning solution, artificial tear, and dry eye prescription that a patient may need to maximize their visual and health outcomes.
There may no longer be a need to analyze toric markers and use calculators for lens rotation (which hopefully you are still doing); instead AI will calculate what parameter changes need to be made. The same holds true for multifocal measurements in terms of selecting add power, zone size, and lens sagittal depth in each eye to maximize the patient’s outcome.
Another utilization of AI may be patients wearing a trial electronic contact lens for a week so that AI would be able to measure their visual needs and calculate what lens currently on the market would best meet their needs.
While AI has been advancing for nearly three-quarters of a century, it is still in its infancy for health care, eye care, and contact lenses. However, new applications for AI are constantly being developed and will be making their way into our lives without us even realizing. Over the next few years, as we further explore AI, we can be excited to keep a vigilant eye on maximizing our patient outcomes with the help of this innovation. CLS
- Ma X, Ahadian S, Liu S, et al. Smart Contact Lenses for Biosensing Applications. Advanced Intelligent Systems. Adv Intel Sys. 2021 May 5. Available at onlinelibrary.wiley.com/doi/full/10.1002/aisy.202000263 . Accessed Oct. 24, 2023.
- Park J, Kim J, Kim S-Y, et al. Soft, smart contact lenses with integrations of wireless circuits, glucose sensors, and displays. Sci Adv. 2018 Jan 24;4:eaap9841.
- Zhang J, Kim K, Kim HJ, et al. Smart soft contact lenses for continuous 24-hour monitoring of intraocular pressure in glaucoma care. Nat Commun. 2022 Sep 20;13(1):5518.
- Leonardi M, Pitchon EM, Bertsch A, Renaud P, Mermoud A. Wireless contact lens sensor for intraocular pressure monitoring: assessment on enucleated pig eyes. Acta Ophthalmol. 2009 Jun;87:433-437.
- Chen G-Z, Chan I-S, Leung LKK, Lam DCC. Soft wearable contact lens sensor for continuous intraocular pressure monitoring. 2014 Sep;36:1134-1139.
- Oliver NS, Toumazou C, Cass AEG, Johnston DG. Glucose sensors: a review of current and emerging technology. Diabetic Med. 2009 Mar;26:197-210.
- Williams N. The real reason google glass failed spectacularly. History Computer. 2023 Aug 10. Available at history-computer.com/the-real-reason-google-glass-failed-spectacularly . Accessed Oct. 24, 2023.
- Efron N. Augmented reality contact lenses–so near yet so far. Clin Exp Optom. 2023 May;106:349-350.