AI Predicts Radiology Workload

AI is touted as a tool that can help radiologists lighten their workload. But what if you could use AI to predict when you’ll need help the most? Researchers in Academic Radiology tried that with an AI algorithm that predicted radiology workload based on three key factors. 

Imaging practices are facing pressure from a variety of forces that include rising imaging volume and workforce shortages, with one recent study documenting a sharp workload increase over the past 10 years.

  • Many industry observers believe AI can assist radiologists in reaching faster diagnoses, or by removing studies most likely to be normal from the worklist based on AI analysis. 

But researchers and vendors are also developing AI algorithms for operational use – arguably where radiology practices need the most help.

  • AI can predict equipment utilization, or even create a virtual twin of a radiology facility where administrators can adjust various factors like staffing to visualize their impact on operations.

In the new study, researchers from Mass General Brigham Hospital developed six machine learning algorithms based on a year of imaging exam volumes from two academic medical centers.

The group entered 707 features into the models, but ultimately settled on three main operational factors that best predicted the next weekday’s imaging workload, in particular for outpatient exams…

  • The current number of unread exams.
  • The number of exams scheduled to be performed after 5 p.m.
  • The number of exams scheduled to be performed the next day.

The algorithm’s predictions were put into clinical use with a Tableau dashboard that pulled data from 5 p.m. to 7 a.m. the following day, computed workload predictions, and output its forecast in an online interface they called “BusyBot.”

  • But if you’re only analyzing three factors, do you really need AI to predict the next day’s workload? 

The authors answered this question by comparing the best-performing AI model to estimates made by radiologists from just looking at EHR data. 

  • Humans either underestimated or overestimated the next day’s volume compared to actual numbers, leading the authors to conclude that AI did a better job of calculating dynamics and weighting variables to produce accurate estimates.

The Takeaway

Using AI to predict the next day’s radiology workload is an intriguing twist on the argument that AI can help make radiologists more efficient. Better yet, this use case helps imagers without requiring them to change the way they work. What’s not to like?

Keeping Pace with Volume: 7 Strategies from ASNR 2025

This week weary neuroradiologists descended upon the City of Brotherly Love for the annual meeting of the American Society of Neuroradiology (ASNR). The field is facing mounting pressure as increasing imaging volumes continue to outstrip radiologists’ capacity. 

Dealing with growing volume was a recurring theme throughout ASNR 2025, with a range of proposed solutions, including the seven strategies below:

  1. Acquisition automation for higher efficiency and reduced technical requirement: A talk by Lawrence Tanenbaum, MD, featured a number of AI solutions to ease technologist training requirements, including smart protocoling, automated patient positioning, one-touch exams without parameter adjustments, and on-device quality assurance and motion correction to cut down repeat exams.
  2. Accelerated acquisitions as the standard-of-care: Every manufacturer – from established vendors to emerging startups – showcased deep learning-based reconstruction. As Suzie Bash, MD, put it, “Deep learning reconstruction is becoming standard-of-care across the industry.”
  3. Improving radiologist reading efficiency with AI and workflow management: A noticeable trend at ASNR 2025 was fewer talks focused solely on algorithm accuracy and more emphasis on how AI impacts reading efficiency. Accuracy remains critical, but adoption increasingly hinges on demonstrating workflow efficiency.
  4. Streamlining new algorithm rollout using integrated platforms: In a session on AI adoption and evaluation, Reza Forghani, MD, PhD, called for increased use of integrated platforms to allow for easier algorithm deployment, validation, and monitoring.
  5. Rising reliance on international medical graduates (IMGs): Mina Hesami, MD, presented on the rising contribution of IMGs to US radiology, noting a steady increase in the proportion of residency slots, fellowships, and leadership roles held by international graduates – with radiology seeing faster growth than most other medical specialties.
  6. Expanding the radiology workforce with mid-level providers: Another proposed strategy is offloading specific tasks to mid-level providers. While still controversial in radiology, this model is gaining traction in response to workforce shortages.
  7. Sustainability by reducing emissions and environmental impact: Several ASNR sessions addressed environmental sustainability. From simply turning off idle scanners to using AI to reduce contrast doses, radiologists are beginning to reckon with the environmental impact of rising scan volumes.

The Takeaway

The sessions at ASNR 2025 indicate that while there’s a lot of buzz around AI, radiologists are considering every tool at their disposal to keep up with rising imaging volumes. AI will play a role, but likely won’t be sufficient alone to keep up with increasing volumes.

T. Campbell Arnold is a research scientist at Subtle Medical and the managing editor of RadAccess.

Imaging Workload Jumps with Higher Use of CT, MRI

Radiology’s shift to more advanced modalities like CT and MRI is increasing the burden on radiologists to interpret more complex exams. A new study in JACR documents the trend, finding that radiologist workload for inpatient imaging has risen sharply over the last 10 years. 

Like many physicians, radiologists are feeling burned out from rising patient workload, personnel shortages, and declining reimbursement. 

  • But radiology has the added burden of being one of healthcare’s most technology-focused specialties, with new imaging modalities giving them cooler tools to work with, but at the cost of steadily increasing exam complexity.

Researchers from Brigham and Women’s Hospital have been tracking inpatient imaging utilization for the past 40 years, and the new paper provides the latest update. 

  • They calculated inpatient imaging volume at Brigham and Women’s from 2012 to 2023, during which 896k imaging exams were performed.  

Results for the study were as follows …

  • Total annual inpatient imaging volume grew 17% over 10 years (102k to 119k exams).
  • Total imaging exams per patient admission (adjusted by case mix and disease severity) fell 20% due to declines in X-ray, ultrasound, and nuclear medicine.
  • But imaging exams per patient admission grew for CT (19%) and MRI (21%).
  • Leading to growth in CT and MRI’s combined share of all radiology global RVUs (62% to 75%).
  • Hospital length of stay rose 32% (5.6 to 7.4 days), possibly due to the COVID-19 pandemic. 

What does it all mean? Basically, the number of inpatient imaging exams per patient is declining when adjusted for disease severity, but radiologists are still having to work harder because the studies are more complex. 

  • Imaging could also be shifting from the inpatient setting to outpatient centers due to reimbursement changes aimed at shifting exams to lower-cost settings than hospitals.

One big question with the new study is the degree to which the COVID-19 pandemic skewed the results compared with previous years. 

  • The pandemic may have spurred more use of CT, especially given its value in providing a definitive diagnosis of SARS-CoV-2 infection. 

The Takeaway

If you feel like you’re working harder than ever, the new findings show that you’re not crazy. And given radiology’s breakneck pace of innovation, it’s not likely the trends revealed in the new study will let up any time soon.

Forecasting Radiologist Supply

Two new studies published this week in JACR raise the provocative question: Will there be a radiologist shortage in the future given growing demand for medical imaging services?

It’s a question that’s become commonplace across healthcare as burnout and other issues prompt many physicians to leave the field. 

  • This has caused workforce shortages that raise questions about whether the U.S. – and other advanced economies – will be able to meet growing demand for healthcare services by an aging population.

The new studies were conducted by Harvey L. Neiman Health Policy Institute researchers and each tackles one aspect of the supply/demand equation over the next 30 years. 

The first study analyzed past growth in the radiologist workforce to find …

  • There were 37.5k radiologists enrolled to provide care to Medicare patients in 2023. 
  • With no growth in the number of residency positions, there will be 47.1k radiologists in 2055, an increase of 26%.
  • If residency positions grow, there will be 52.6k radiologists, an increase of 40%.

The wildcard here is growth in residency positions, which are mostly controlled by Medicare through its graduate medical education program – and it literally takes an act of Congress to increase the number of trainee positions. 

  • Another factor is whether the higher physician attrition rate seen during the COVID-19 pandemic continues into the future. 

The second study addressed growth in imaging volume by analyzing trends in claims data for Medicare, Medicaid, and private insurance, finding …

  • Imaging utilization will be 17-27% higher by modality by 2055 assuming no continuation of recent utilization trends.
  • Most utilization growth will be seen in nuclear medicine (27%), CT (25%), interventional radiology (23%), X-ray (18%), and MRI and ultrasound (17% each).
  • Adding recent utilization trends to the model finds utilization by 2055 either -5.6% lower or up by 45%.

Factors affecting future utilization include population growth (73-88% of increase) and population aging (12-27%). 

The Takeaway

So will there be a radiologist shortage in the future? The new studies indicate that there are too many variables to make an accurate prediction right now. But they do provide a foundation for future research – and debate. 

Get every issue of The Imaging Wire, delivered right to your inbox.

You might also like..

Select All

You're signed up!

It's great to have you as a reader. Check your inbox for a welcome email.

-- The Imaging Wire team

You're all set!