-
Artificial intelligence could make radiology reports for X-ray, CT and MRI scans twice as understandable without compromising clinical accuracy, according to new research from the University of Sheffield.
-
Researchers found that when AI systems such as Chat GPT were used to rewrite these reports, reading levels dropped from college level to those of 11- to 13-year-old students.
-
This could have significant benefits for patients and health systems, reducing anxiety and confusion and improving health equity for people with low health literacy and those who use English as a second language. It also frees up clinician time to focus on treatment and care decisions.
A major new study from the University of Sheffield suggests that artificial intelligence could help patients understand complex medical scan results, making them much easier to understand without compromising clinical accuracy.
The study found that when radiology reports for X-ray, CT, and MRI scans were rewritten using advanced AI systems such as ChatGPT, they were almost twice as easy for patients to understand compared to the original versions.
The analysis showed that reading levels had dropped from “college level” to a level closer to that of 11- to 13-year-old students.
The findings suggest that AI-assisted explanations could become a standard adjunct to medical reports, helping to improve transparency and trust across health systems, including the NHS.
Researchers reviewed 38 studies published between 2022 and 2025, covering more than 12,000 radiology reports simplified using AI. These rewritten reports were evaluated by patients, the public, and clinicians to assess both patient understanding and clinical accuracy.
Radiology reports are traditionally written for physicians rather than patients. However, initiatives to promote patient-centred care, such as the NHS app, and new policies mandating greater transparency in medical records, have rapidly expanded patient access to these reports.
The study’s lead author, Dr Summer Arabedo, Senior Clinical Research Fellow at the University of Sheffield and Honorary Consultant Cardiac Radiologist at Sheffield Teaching Hospitals NHS Foundation Trust, said: “The fundamental problem with these reports is that they are not written with patients in mind. They are full of jargon and abbreviations that can be easily misunderstood, causing unnecessary anxiety, a false sense of security and confusion.”
“Patients with low health literacy or who speak English as a second language are particularly disadvantaged. Clinicians often have to spend valuable appointment time explaining reporting terminology instead of focusing on care and treatment. Even small time savings per patient can have significant benefits across the NHS.”
Physicians who reviewed these AI-simplified reports found that the majority were accurate and complete, but about 1% contained errors, such as incorrect diagnoses. We show that while this approach is very promising, it still requires careful monitoring.
Of the 38 studies reviewed, none were conducted in the UK or NHS settings, and Dr Sommer said the research team was now trying to address this major gap.
“This study has highlighted several important priorities, most importantly the need for real-world testing in NHS clinical workflows to properly assess safety, efficiency and patient outcomes,” Dr Sommer said.
“This includes a human oversight model in which clinicians review and approve AI-generated instructions before sharing them with patients. Our long-term goal is not to replace clinicians, but to support clearer, kinder, and fairer communication in healthcare.”
This research highlights the university’s ambition to transform ideas into impact, a true embodiment of independent thinking and shared ambition.
Click here to read the full research paper.
/Open to the public. This material from the original organization/author may be of a contemporary nature and has been edited for clarity, style, and length. Mirage.News does not take any institutional position or position, and all views, positions, and conclusions expressed herein are those of the authors alone. Read the full text here.
