When Patients Read Radiology Reports First

Could the era of the empowered patient change how imaging reports are structured and presented? It’s a question on the mind of healthcare leaders and solution providers as patients begin asking artificial intelligence (AI) models for assistance in interpreting radiology studies.

Research published in the American Journal of Roentgenology shows that 44% of radiology reports are now viewed by the patient before the ordering doctor has seen them.¹ Those reports, however, include clinical language, differential diagnoses, incremental findings, and in some cases, ambiguity, all of which makes them difficult for a patient to comprehend without additional interpretation.

“The report is written for a purpose, to help a medical expert take the next step,” explains Morris Panner, president of Intelerad, now a part of GE HealthCare. “But as a patient, the main concern is, ‘Am I OK?’”

Panner recently experienced this dichotomy firsthand when he took the results of a personal imaging scan, loaded them into an AI engine and asked the large language model (LLM) to interpret the report. “It told me, in general, it looks like you’re fine, although there’s a question or two I might want to ask a doctor,” he says.

Who Should Radiology Patients Consult?

Panner’s choice to use AI was based on convenience for both him and his provider. “I knew my physician was probably triaging and saying that if I’m basically OK, that’s pretty far down the list,” he says. “Why not just spare them that kind of additional pressure, which is already adding to their impossible day?”

Panner is far from alone. ChatGPT fields health and wellness questions from 230 million patients worldwide every week.² Research published earlier this year validates that LLM-generated, patient-friendly summaries increase comprehension and reduce anxiety, representing a potential tool to improve communication.³

Some patients, however, want to hear results from a physician first. The question is, should that doctor be the ordering physician or the radiologist? “That’s one of those topics where, if I go to dinner with five radiologists, I can start a pretty great debate,” Panner says.

He notes that leaders like Keith Hentel, MD, executive vice chair of the Department of Radiology for NewYork-Presbyterian Weill Cornell Medical Center, champion radiologists talking directly with patients.

“Others aren’t as open to the idea because — and I understand this perspective, too — radiologists are the ‘doctor’s doctor.’ They’re giving very specific guidance and a very technical set of decisions [to ordering doctors].”

Making Reports Work for Clinicians and Patients

As solution providers such as Intelerad embed more AI features into their platforms, the challenge is doing so in ways that promote patient-centered radiology while also reducing the time burden radiologists face.

“Radiologists tell me that their ‘product’ is the diagnosis that goes out under their name,” Panner says. “I signed this report, that is my diagnosis, my professional judgment, my reputation, my life on the line. So what you have isn’t a question of a change of output. You have change of access and change of collaboration.”

Enabling that communication will require two steps. First, Panner says, is embedding multimodal reporting and patient-friendly summaries into enterprise tools. “You have a lot of reporting systems that make the report easier to generate for the radiologists and give whatever they want,” he says. “So you can take a report and have it be multimodal, where you can have a summary that’s very patient friendly, and you can specify how you want that to work.”

Second, Panner sees next-generation, digital-first workflows bringing benefits to both patients and providers, especially when it comes to gathering input from multiple specialists efficiently in diagnosing and treating tumors or stroke. “A digital report that can be easily disseminated, can live in a lot of different places, can really help streamline care,” he says.

At all steps, Panner adds, the radiologist must be kept in the loop to protect patient safety. “I see AI as another way to make humans more powerful, which is great,” he says.

References

1. Pollock JR, Petty SAB, Schmitz JJ, Varner J, Metcalfe AM, Tan N. Patient Access of Their Radiology Reports Before and After Implementation of 21st Century Cures Act Information-Blocking Provisions at a Large Multicampus Health System. AJR Am J Roentgenol. 2024;222(6):e2330343. doi:10.2214/AJR.23.30343

2. OpenAI. Introducing ChatGPT Health. OpenAI. Jan. 7. Accessed July 23. https://openai.com/index/introducing-chatgpt-health

3. Serna JA, Yu Y, Diaz P, Sakiyama K, Ye M, Rustagi A, Sohn JH. Self-reported comprehension of large language model-generated summaries of lung cancer screening reports: a vignette survey. Radiology Advances. 2026;3(2):umag008. doi:10.1093/radadv/umag008