Every three to four seconds. That's how often the average radiologist must interpret a single image during an eight-hour shift. This pace, when sustained day after day, produces something closer to an endurance sport than a medical practice. The result is a profession that’s burned out, understaffed, and increasingly reliant on technology like AI that has yet to close the gap.

The data tells a stark story. According to Radiology Unlocked: The Global Radiologist Report 2025, more than half of all radiologists name burnout as their single greatest professional concern, edging out the workforce shortage. A survey published in Diagnostic Imaging found that 44% of male radiologists and 65% of female radiologists report experiencing burnout, with some studies placing the overall figure above 80%. These are not statistics about dissatisfied employees. They are warning signs of a healthcare system stretched to its structural limits.

A Shortage That Shows No Sign of Easing

The United States currently has approximately 37,500 practicing radiologists, roughly 1,500 fewer than the field requires. Without significant intervention, that deficit is projected to more than double to 3,100 in the near future, according to research published in HealthManagement. Meanwhile, the pipeline into the profession remains narrow: only 29 new radiology residency training spots were added between 2021 and 2025.

The demand side of the equation is accelerating in the opposite direction. An aging population, with an estimated 20% of Americans now on Medicare, combined with the rise of chronic disease, is driving volumes ever higher. 

Some hospital systems have already begun making painful concessions. Certain outpatient imaging centers have been temporarily shuttered so that radiologists can address growing backlogs elsewhere. Reports have emerged of radiologists experiencing exhaustion and distress severe enough to prompt thoughts of suicide. Overwhelmed radiologists leave or reduce hours, increasing pressure on those who remain, which accelerates further departures.

The American College of Radiology has estimated that burnout-related turnover costs health systems an average of $500,000 per departing radiologist once recruitment, training, and productivity losses are factored in.

AI: Promise, Paradox, and Limits

AI has been heralded as the technology that would either replace radiologists entirely or save them from the demands consuming their profession. The reality is more nuanced.

On the capability side, AI has made genuine strides. Hundreds of models have received FDA approval for tasks ranging from pneumonia detection to breast cancer screening. A review published in Health Technology found that AI can interpret images faster than radiologists and could meaningfully reduce the practical impact of the global shortage. Some estimates suggest AI could handle up to 53% of current radiologist workloads.

Despite these advances, the radiologist shortage hasn’t abated. Radiology residency programs are not keeping pace with demand. Salaries are climbing as competition for talent intensifies. The shortage is not a problem AI has solved; it’s a problem AI has not yet been able to meaningfully address at the structural level.

Perhaps more surprising is what recent research reveals about AI's relationship with the burnout it was supposed to cure. A study published in JAMA Network Open found that AI use was associated with increased burnout among radiologists, meaning the more tools a radiologist used, the more burnout they reported. This was especially pronounced among radiologists with already heavy workloads and those with low acceptance of the technology. Rather than relieving burden, poorly integrated AI added friction to an already overloaded workflow.

The lesson that emerges from the research is not that AI is the wrong answer. It’s that AI, applied haphazardly or without clinical workflow integration, can be part of the problem. 

It’s Not Just the Volume

The burnout in radiology is not simply about volume. It’s about the accumulated cognitive and administrative drag that turns a skilled clinician into a system administrator. Consider the typical radiology shift: a physician arrives and must navigate multiple separate worklists across different PACS, manually triage incoming studies, manage uneven caseload distribution, and spend meaningful portions of their shift on tasks that have nothing to do with diagnosis.

Where Medicom Fits

This is the challenge Medicom was built to address. While much of the conversation around AI focuses on diagnostic algorithms, our approach targets the upstream obstacles that compound every radiologist's day before a single image is interpreted.

By standardizing medical image exchange on a single, unified platform, Medicom helps eliminate the fragmented patchwork of VPNs, physical media, and disconnected systems that forces radiologists to spend clinical hours on logistics rather than diagnosis. One of our health system clients noted that Medicom "simplified our internal workflows while managing the complexity of a large external network," allowing the system to re-engineer its image sharing model to scale with growth while staying focused on patient care.

The Road Ahead

The radiology workforce crisis will not resolve quickly. Structural changes, including expanding residency slots, updating immigration pathways for foreign-trained radiologists, and rethinking care team models with radiologist assistants, will take years to materialize at scale. AI will continue to mature and, when thoughtfully implemented, can meaningfully reduce the burden on radiologists. But neither technology nor policy alone will cure what is fundamentally a systemic imbalance between supply and demand.

The tools radiologists use every day can be better, and the improvements are not complicated. Seamless image sharing, reduced administrative burden, and smarter worklists would help close the gap between what radiologists could accomplish with ideal infrastructure and what they actually accomplish while navigating broken workflows. That gap is where the most immediate opportunity lives, and where the most preventable burnout originates.

Radiologists are not exhausted because the work is too hard. It’s the systems surrounding the work that make it more difficult. Solving the shortage may be a decade-long project, but solving the friction is a choice that can happen today.

Curious about how our clients are turning fragmented imaging into a unified workflow? We'd like to compare notes.