Clinical Judgment
When AI Recommends Your Lens... But Your Eye Tells a Different Story
AI can help you research. It cannot examine your eyes. A real case that shows why the conversation still matters.
He came prepared. More prepared than most patients I see.
A 72-year-old man walked into my clinic for a cataract consultation carrying a folder filled with research. He had done what many patients are now doing: he had consulted multiple AI tools to help guide his decision.
Interestingly, they all gave him the same answer: a specific type of advanced lens called an Extended Depth of Focus lens.
On paper, it made sense.
But medicine is rarely just "on paper."
When I examined his eyes, I found a few important details that no AI tool could have known:
Called an epiretinal membrane. It is a thin layer of scar tissue on the surface of the macula. Even a mild one can affect contrast sensitivity and fine visual quality, especially with premium lens designs that rely on precise light distribution.
From years of contact lens wear. These surface changes can scatter light in ways that degrade the performance of advanced optics. A standard exam might miss them. Detailed topography does not.
These are the kinds of nuances that can affect how well certain premium lenses perform, especially when it comes to contrast and clarity.
We had a long conversation. Not the kind where I tell someone what to do. The kind where I lay out what I see, what it means, and what the realistic range of outcomes looks like.
I explained that while the lens he chose is excellent and FDA-approved, patients with even mild retinal changes do not always get the same crisp quality of vision. I wanted him to understand that he might not achieve perfect 20/20.
To his credit, he listened carefully.
And then he made a thoughtful decision: he still wanted to move forward.
Not because he ignored what I told him. Because he understood the tradeoffs and decided the potential benefit was worth it for his life.
That is informed consent at its best.
Two weeks after surgery, he came back.
His vision measured 20/25. Not perfect on the chart.
But his reaction told the real story.
He was happy.
Before surgery, he was 20/50 and dependent on contact lenses. Now he could see clearly enough to drive, use his phone, and work on his computer, all without contacts.
That was a meaningful improvement in his life.
Success is not always about achieving "perfect" vision. It is about achieving the vision that works for your life.
AI is becoming a powerful tool for patients. It can help you learn, explore options, and feel more involved in your care. I am genuinely glad patients are doing their research.
But it has limits.
It cannot feel the texture of your cornea or see the subtle membrane on your retina.
No questionnaire replaces a slit lamp exam, topography, and OCT imaging.
Medicine at its best integrates data with judgment, experience, and the ability to read the human being sitting across from you.
The best decisions happen when your research and your doctor's expertise come together. Not in competition, but in collaboration.
Come prepared. Do your research. Ask AI anything you want. But then sit down with a surgeon who will actually look at your eyes, tell you what they find, and help you make a decision that accounts for the things no algorithm can see. That is where the best outcomes live: at the intersection of your questions and your doctor's experience.