Digital pathology unlocks remote sign-out, subspecialty consultation across institutions, AI-assisted review, faster turnaround times, better collaboration on complex cases, and quantitative analysis that glass slides simply cannot support. Those are just a few of the benefits. The case for going digital is strong. The question is whether pathology can capture those gains without inheriting the workflow debt radiology is still paying off.
Radiology went digital three decades before pathology. That head start is often translated into a corresponding technical roadmap for digital pathology: PACS, DICOM, worklists, AI. However, the more important lesson stems from the experience of the people inside those workflows. Roughly 61% of radiologists report burnout, up from 36% in 2013 and 49% in 2017. Practice leaders agree it is a serious problem: 77% say burnout is significant or very significant in their group.1 PACS is not the sole cause, but it is consistently named as a contributor. The American College of Radiology’s 2018 statement on radiologist well-being lists PACS and the electronic medical record (EMR) alongside productivity demands and isolation as structural risk factors.2
Going digital didn’t cause burnout, going digital without redesigning the workflow did.
PACS solved real problems. Films stopped getting lost. Priors became available in seconds. Reads could be distributed across sites and time zones. These were genuine gains, and nobody wants them reversed.
PACS also removed the physical constraints that had quietly governed radiologist workload. Once access was frictionless, volume expanded to fill the new capacity. Between 1991 and 2007, annual RVUs per full-time radiologist increased by roughly 70%.3 The stakes around individual decisions have sharpened to the point that malpractice cases now turn on keystroke timing data: a 2020 case alleging a missed subdural hematoma depended on subpoenaed PACS logs showing the exact seconds the radiologist spent on a head and cervical spine CT.4 That is not a sustainable operating point, but it is not a PACS problem; it is a workflow and incentive problem that PACS enables.
Remote access also had a cost that rarely shows up in productivity reporting: isolation. PACS let radiologists read from anywhere, which eliminated most of the incidental hallway contact with referring clinicians. That loss of face-to-face consultation is one of the most consistently cited drivers of radiologist burnout in the literature, producing depersonalisation and a diminished sense of meaning in the work. The specialty got faster and more scalable, while simultaneously detaching from the rest of medicine.
Pathology is not starting from zero on burnout.
One might assume that pathology is insulated because the work has historically been slower-paced and collegial. The data does not support that. A 2022 survey of Canadian pathologists using the Maslach Burnout Inventory found 58% met criteria for burnout, higher than previously reported baselines for US pathologists (roughly 33–45% across ASCP and Shanafelt studies).5 A separate 2023 pre-pandemic study across pathology and laboratory medicine found a 58.4% burnout rate, with workload, control, and loss of meaning in work ranking as top drivers.6 Workload is consistently the number one source of job stress.7
In other words: pathology is entering its digital transition with burnout rates already comparable to where radiology is now, and with the same structural drivers sitting at the top of the list. If digital pathology follows the radiology template without adjustment, those numbers will not improve. They’ll get worse.
What the radiology experience actually teaches.
Volume expectations follow capacity, unless leadership holds the line.
The single biggest mechanism behind radiology’s burnout trajectory was that productivity targets rescaled to match what digital made possible. Pathology groups adopting whole slide imaging and AI-assisted review should define what a sustainable caseload looks like before scale becomes the expectation, not after. This is a leadership and compensation design question, not a technology one.
The viewer and workflow have to be redesigned together.
Radiology’s measurable wins in turnaround time, collaboration, and access did not come from the monitor. They came from worklist logic, case routing, integration with priors, and reporting workflow. But the viewer is where those workflows either come together or fall apart. A viewer designed only to display images, with worklists, case context, ancillary data, and reporting bolted on around it, forces the pathologist to be the integration layer. That is how digital pathology reproduces the old workflow at higher speed, which is exactly the failure mode to avoid. The viewer has to be built as a workflow surface from the start: case assembly, synoptic reporting, ancillary integration, and handoffs all live inside the same environment the pathologist is already working in.
Measure the right things early.
Adoption metrics, such as percent digital sign-out and slide volume, are vanity. The metrics that predict sustainability are turnaround time by workflow step, rescan and QC rates, queue health, time-in-tool, and staff-reported usability and control. Radiology departments that tracked operational and human factors such as workflow bottlenecks, turnaround time by step, and user experience adapted to digital workflows more effectively than those focused primarily on throughput or volume metrics, as demonstrated in PACS adoption, workflow optimisation, and clinician burnout literature.8, 9
AI should remove toil before it adds capability.
AI in radiology delivers the greatest practitioner benefit when it absorbs repetitive or preparatory work, such as measurement, prior-study comparison, triage, and report pre-population. It delivers the least benefit when it layers new review steps on top of existing workflows. This will also be true in pathology. AI that absorbs preparatory work, including region of interest prescreening, quantification, and case prioritisation, returns time and attention to the pathologist. A model that flags findings the pathologist must then adjudicate outside their existing workflow does the opposite: it adds a new layer of interruption without removing anything.
The advantage pathology actually has.
Radiology learned these lessons in real time, under volume pressure, with legacy systems. Pathology can learn them from history. The technical groundwork, including interoperability standards, cloud infrastructure, and validated AI, is more mature than PACS was in 1995. The operational choices that govern how work is organised, measured, and compensated around the digital case will determine whether the next decade of pathology diverges from radiology’s history or repeats it.
Build for what comes next.
Pathology has the chance to apply radiology’s lessons from day one. This requires three things: the viewer and workflow must be designed as one environment, the metrics must capture both operational and human signals, and AI must be integrated into the workflow rather than layered on top of it.
Techcyte Fusion® was built with these three principles at its core. Case assembly is where it begins. Fusion goes further by integrating deeply with both the LIS and the EHR, so clinical context, prior cases, and concurrent cases are in the same environment the pathologist is already working in, alongside synoptic reporting, ancillary test results, and consults with subspecialists. The viewer is a workflow surface, not a display.
AI is integrated in the same way. Rather than sitting as a separate tool the pathologist has to open, check, and reconcile, AI runs within the workflow itself, aligned to how cases progress through the lab and surfacing results where and when the pathologist is already focused. Operational and usability metrics are core features, not add-ons. None of that makes digital transformation automatic, but it does create the conditions to scale both performance and the people delivering it.
Selecting a platform for digital pathology is ultimately a choice about what to scale, and an opportunity to build a future where efficiency, quality, and sustainability advance together.
Sources:
- GE HealthCare, “4 Ways Technology Can Help Reduce Radiologist Burnout” (2022)
- Spieler B, Baum N. “Burnout: A Mindful Framework for the Radiologist.” Current Problems in Diagnostic Radiology (2021)
- Chetlen AL et al. “Addressing Burnout in Radiologists.” Academic Radiology (2019)
- Alexander R, Waite S, Bruno MA et al. “Mandating Limits on Workload, Duty, and Speed in Radiology.” Radiology (2022)
- “The Burnout in Canadian Pathology Initiative.” Archives of Pathology & Laboratory Medicine (2023)
- Smith SM et al. “Burnout and Disengagement in Pathology.” Archives of Pathology & Laboratory Medicine (2023)
- “The American Society for Clinical Pathology’s Job Satisfaction, Well-Being, and Burnout Survey of Pathologists.” American Journal of Clinical Pathology (2020)
- “The value of workflow analysis in radiology.” Journal of Digital Imaging, Reiner BI, Siegel EL.
- “What the radiologist should know about artificial intelligence – an ESR white paper”, European Society of Radiology, Insights into Imaging, (2019)