Can AI Write Radiology Reports? A Study Suggests It Can — and Save Time Doing It

  • Home
  • Newsroom
  • Can AI Write Radiology Reports? A Study Suggests It Can — and Save Time Doing It
Can AI Write Radiology Reports? A Study Suggests It Can — and Save Time Doing It - Axcellant

Can AI Write Radiology Reports? A Study Suggests It Can — and Save Time Doing It

  1. Aug 21, 2025

A recent publication in npj Digital Medicine explored the use of AI-generated radiology reports — and the results are worth watching.

In the study, radiologists used a model that generated full reports from short keyword prompts. The result? A nearly 30% reduction in reporting time, without compromising quality or accuracy. The goal: to improve efficiency and reduce variability in radiology workflows.

For clinical trials, especially those involving high volumes of imaging data, this approach has real potential. Consistency in documentation and faster turnaround times could benefit sponsors and CROs managing imaging-heavy protocols — while still keeping expert oversight in place.

At Axcellant, we believe AI won’t replace clinical experts, but it will increasingly support their work in subtle, meaningful ways — especially in high-pressure, data-intensive environments.

Read the full study here.

Resources

Large Language Models in Nuclear Cardiology: From Experimental Tools to Clinical Infrastructure

3 Questions with: Prof. Piotr Słomka Large language models are entering nuclear cardiology at a pace that makes the usual…

From Half-Life to Clinical Impact: Mastering Time-Critical Supply Chains

How Axcellant helped a diagnostic study overcome isotope supply, manufacturing, and logistics constraints In radiopharmaceutical development, we often say that…

How ACT EU Is Forcing a Global Approach to Trial Timelines

Europe’s clinical trials market is entering a more practical phase in its approach to competitiveness. With the Clinical Trials Regulation…