Solving Radiology Bottlenecks Through AI
AI-Powered Solution Boosts Productivity by 40% Without Compromising Accuracy
The good news: X-rays and CT scans help physicians find health problems quickly and without surgery.
The challenge: imaging volumes continue to rise, driven by an aging population and the growing prevalence of chronic diseases in the United States.
This increasing volume expands more quickly than the radiology workforce, which leads to a strain on radiology services. This imbalance between supply and demand creates a bottleneck — resulting in longer turnaround times, which could delay a diagnosis. It also puts a lot of strain on the existing workforce, which contributes to burnout.
To address this, Northwestern Medicine engineers built their own generative AI tool from scratch, using data to create a tool specifically designed for radiology.
ARIES: A Second Pair of Eyes
Unlike other AI tools currently on the market, ARIES uses a holistic model that analyzes the entire X-ray or CT scan (Figure 1). It then generates a personalized report that is 95% complete. It summarizes the findings in the radiologist’s reporting style and offers a template to augment the radiologist’s diagnosis and treatment. The system also flags life-threatening conditions in real time.
Figure 1: ARIES generates a report using an image and input, allowing radiologists to review in a timelier manner.
For the study, ARIES was deployed across all Northwestern Medicine locations, and it analyzed nearly 24,000 radiology reports over a five-month period in 2024. The team then reviewed the creation times and clinical accuracy with and without the tool.
The results:
In other words, significant time was saved while ARIES maintained the clinical accuracy and textual quality of traditional reports.
“It’s not an exaggeration to say that it doubled our efficiency,” says Samir Abboud, MD, chief of Emergency Radiology at Northwestern Medicine and co-author on the study. “It’s such a tremendous advantage and force multiplier.”
It’s not an exaggeration to say that it doubled our efficiency
Samir Abboud, MD, Chief of Emergency Radiology at Northwestern Medicine
Time Saved Means Faster Diagnosis
ARIES also helps clinicians with worklist prioritization.
“On any given day in the ER, we might have 100 images to review, and we don’t know which one holds a diagnosis that could save a life,” says Dr. Abboud. “This technology helps us triage faster, so we catch the most urgent cases sooner and get patients treatment quicker.”
The time saved allows radiologists to return diagnoses much faster, particularly in critical cases where every second counts. Several scans in the study found incidental findings, such as lung nodules or collapsed lungs.
This technology helps us triage faster, so we catch the most urgent cases sooner and get patients treatment quicker.
Samir Abboud, MD, Chief of Emergency Radiology at Northwestern Medicine
Looking Ahead
ARIES will expand to support MRI and ultrasound imaging. The team is also adapting the AI model to detect potentially missed diagnoses, such as early stage lung cancer.
Four patents have been approved for the ARIES technology, with more to come.
