AI Chest X-ray triage of normal / abnormal boosts radiology efficiency in NHS

In a recent pilot study at Frimley Health NHS Foundation Trust, Qure.ai’s AI-driven radiology tool, qXR, has demonstrated significant potential to ease the workload of consultant radiologists. The study aimed to assess qXR's ability to categorize chest X-rays (CXRs) as normal or abnormal, targeting a portion of the caseload—mainly GP and outpatient requests—which typically consists of 40% normal CXRs. By automating this triage process, the AI tool could redirect normal cases to radiographers, freeing up consultants to focus on complex imaging work.

The early results are promising: qXR showed a remarkable 99.7% accuracy in identifying normal CXRs. This efficiency translates to a potential reduction of up to 58% in radiologists' caseloads, saving them as much as two hours per day. With this time, radiologists can dedicate more attention to intricate diagnostic work, enhancing the overall effectiveness and responsiveness of radiology services.

Beyond workload management, qXR’s impact on patient care is noteworthy. The AI tool successfully flagged all cancer cases, including subtle nodules that might otherwise be overlooked. This ability to identify high-risk, early-stage lung cancer cases is a breakthrough, supporting earlier diagnosis and treatment, which are crucial for improving survival rates.

Dr. Amrita Kumar, a Consultant Radiologist and AI Clinical Lead at Frimley Health, highlighted the transformative role of AI, noting how it enables radiologists to prioritize complex cases while streamlining routine workflows. Darren Stephens from Qure.ai echoed this sentiment, emphasizing AI’s role in addressing the ongoing radiologist shortage, projected to reach 40% by 2027 according to the Royal College of Radiologists. As demand for radiology services grows, this study illustrates the promising potential of AI like qXR in relieving pressure on human resources, ensuring efficient and high-quality patient care across the healthcare system.

Frimley Health’s pilot of qXR marks a step forward in integrating AI into clinical settings, showcasing how innovative technology can address resource constraints while advancing patient outcomes.