For years, the debate about AI in radiology has been framed in the broadest possible terms. Will it replace radiologists? Will it surpass the specialists? Will it make diagnostic imaging faster, cheaper and more accurate?
Those questions matter. But they can distract us from a more immediate and practical opportunity. The first truly scalable use of AI in imaging may not be autonomous diagnosis. It can help patients understand the imaging information they already receive.
This is a bigger problem than many healthcare leaders realize.
Increasingly, patients see results before speaking to a doctor. The 21st Century Cures Act accelerated that change by requiring immediate electronic access to many types of health information. in a survey of 8,139 patients96% said they wanted to continue receiving test results posted immediately online, even if a doctor had not yet reviewed them. This is a powerful vote in favor of transparency. But transparency alone does not create understanding.
Radiology reports were never written for lay readers. TO study found that only a very small proportion of radiology reports in a large American health system were readable at the eighth-grade level, which is about the reading level of the average American adult. The rest was filled with dense terminology, acronyms, sequence names, and compressed clinical reasoning. This can work for communication between specialists. It doesn’t work as well for patients trying to understand what’s going on inside their own bodies.
CT and MRI make this challenge even more difficult. These reports often combine anatomy, technical imaging language, incidental findings, and a differential diagnosis in a format that assumes medical training. For many patients, the result is not clarity but confusion.
None of this happens in an emotional vacuum. Waiting for imaging results is stressful, especially in oncology, neurology, and other high-risk settings where a scan can determine an important treatment decision. One study found that patients typically expected outpatient imaging results within one to three days, and 45% reported an emotional change while waiting, most commonly anxiety. In practice, the patient experience is not simply access to results. It is access to complex results, under stress, often before the explanation arrives.
That is the context in which patient-oriented imaging AI becomes important.
What has changed is not that AI is now ready to diagnose patients independently. It’s not. But open medical models are becoming capable enough to act as translation layers between technical imaging data and patient understanding. from google MedGemma 1.5 model card It’s notable here because it expands support for interpreting three-dimensional volume representations of CT and MRI, not just 2D medical images. You can also generate text output such as responses, image analysis, and summaries. In benchmarks published by Google, MedGemma 1.5 4B improved over previous versions on internal CT and MRI classification tasks.
This is a significant change. It suggests that medical AI is moving beyond generic chatbot behavior toward supporting modality-aware imaging.
This does not mean that AI is ready to replace radiologists, or that it should. In fact, the more useful question is another: should healthcare organizations create a layer of patient explanation around the images, with appropriate guardrails?
If used well, that layer could do several things that the current system doesn’t do well. I could translate the jargon into simple language. Could you explain what a sequence is and why it is important. You could answer follow-up questions about anatomy, report structure, and common next steps. It could help patients prepare for more informed conversations with their doctors. Most importantly, it could reach patients at the exact moment when portals provide information but workflows do not yet provide context.
That’s important for doctors, too. Immediate publication of test results has implications for workflow. in one analysisDaily patient messages sent within six hours of reviewing results increased substantially after transitioning to Cures Act compliance. If health systems are going to continue opening the door to raw results, they also need to invest in the explanatory infrastructure behind that door.
The key is to use AI in the right role. Healthcare does not need to hand over diagnosis to AI for this to be valuable. A safer approach is to use AI as a patient education layer that sits alongside the official radiology report, not instead of it. That means telling patients clearly when the system is uncertain, avoiding treatment advice, flagging findings that need human follow-up, and reviewing how the tool performs in the limited role it’s meant to play: helping people understand their images, not making clinical decisions for them.
Healthcare has spent too much time wondering whether AI can read scans like a doctor. The most pressing question is whether you can help patients understand what your portal already shows them.
That’s not a minor problem. It’s what patients feel first.
About Peter Nemeth
Peter Nemeth is the founder of ReadYourLab.comEnterprise software architect and cancer survivor. His work focuses on how AI can help patients better understand complex CT and MRI findings and prepare for more informed conversations with their doctors.
