Why We Stay: Challenge, Mastery, and Passion in Imaging Informatics
I came to radiology knowing computers and nothing about medicine. More than 16 years later, I am still here. This is my attempt to explain why.

In this article8
There is something unusual about the people who truly fall in love with imaging informatics.
They learn its systems far beyond what their job descriptions require. They experiment, automate, troubleshoot, and continuously look for ways to make workflows better. A problem that might cause someone else to open a support ticket becomes a challenge they want to understand and solve themselves.
After more than 16 years working in radiology informatics, I have realized that I am one of those people.
I entered the field without a background in medicine or anatomy. I had never seen an MRI outside of Hollywood productions. I did not know the organs of the abdomen or understand how the different imaging modalities worked. I knew computers, and I never expected radiology itself to become something I cared about.
Over time, however, learning the technology required learning radiology itself. PACS led to DICOM. DICOM led to modalities, routing, worklists, image presentation, and workflow. Interfaces required understanding orders, results, patient information, and the systems surrounding imaging. Supporting radiologists required understanding not only whether technology worked, but how they used it and why seemingly small workflow details mattered.
What began as an average IT job became something much larger.
That experience has led me to a question that I believe is worth asking:
What is it about imaging informatics that creates such deep attachment among the people who truly fall in love with it?
I believe part of the answer is surprisingly simple: imaging informatics provides an almost endless supply of challenges and, with them, an endless supply of rewards.
An Unfamiliar World
Most people have at least a basic understanding of banking, retail, accounting, manufacturing, or other common industries before supporting their technology. They may not understand the specialized systems involved, but the underlying concepts are familiar.
Medical imaging is different.
Most people know what an X-ray, CT, or MRI is, but comparatively few people outside of healthcare have spent meaningful time looking through imaging studies or understanding how those images are produced, transmitted, stored, interpreted, and ultimately turned into information used for patient care.
For someone entering imaging informatics without a clinical background, there are therefore two disciplines to learn simultaneously. There is the technology: servers, networks, databases, applications, interfaces, standards, storage, security, and increasingly cloud infrastructure and artificial intelligence. Then there is radiology itself: modalities, anatomy, examinations, protocols, image series, measurements, priors, reporting, radiologist workflow, technologist workflow, and the clinical importance of getting all of it right.
Eventually, the boundary between the two begins to disappear.
A good imaging informatics professional does not need to be a radiologist, but understanding the technology increasingly requires understanding the people using it and the clinical work the technology exists to support.
That creates an unusually deep technical domain to explore.
The Art Did Not Disappear. It Became Digital.
There was an art to operating radiology before PACS.
Images existed on film. They had to be developed, handled, organized, transported, retrieved, and correctly presented. Prior examinations had to be located and physically made available. An enormous amount of work occurred behind the scenes before a radiologist could interpret an examination.
Modern technology did not eliminate that orchestration. It digitized it.
Today, an exam may begin with an order originating in one system. Patient and procedure information must reach the modality. The modality produces images using technology that is itself extraordinarily complex. Those images must be correctly identified, transmitted, routed, stored, and made available to the appropriate radiologist. Prior examinations may need to be identified and retrieved from another archive or facility. The radiologist needs the correct images, clinical information, tools, hanging protocols, reporting system, and workflow context. The resulting report then has to travel back through another collection of systems to the people responsible for the patient’s care.
When everything works correctly, the process can appear almost effortless.
It is not.
Behind that experience may be PACS, RIS and EHR systems, DICOM, HL7 or FHIR interfaces, databases, networks, VPNs, routing engines, archives, dictation systems, worklists, cloud services, artificial intelligence, and countless configuration decisions made over many years. Perhaps one of the greatest accomplishments in radiology informatics is making all of that complexity invisible.
The goal is often to make something difficult look easy.
The Challenge Is the Reward
Early in my career, I noticed something about the way I approached technology: I wanted to understand it.
When I encountered a repetitive task, I wanted to automate it. When I encountered a system I did not understand, I wanted to master it. When someone told me that something could not be done, my first reaction was usually to determine whether they were actually correct. Often, being told something was impossible made me want to solve it more.
One of the PACS platforms I supported early in my career relied heavily on configuration files containing an enormous number of poorly documented settings. I made it a personal goal to understand those settings, and that process took years. I learned by observing vendor support, experimenting, applying knowledge of the application, remembering previous solutions, and occasionally discovering that assumptions about what the system could or could not do were wrong.
A support ticket might solve a problem once. Understanding the problem, and the technology, could solve it forever.
More importantly, every problem solved increased my ability to solve the next one.
That creates a cycle:
ChallengeCuriosityExperimentationSolutionMasteryNew Challenge
For someone whose mind responds strongly to problem solving, that cycle can become remarkably rewarding.
I have never truly run out of things to learn or problems to solve. Even when nothing is broken, there is always something that could be better. There is a repetitive process that could be automated. A workflow contains unnecessary steps. Two systems exchange information poorly. Data exists but is difficult to use. A radiologist performs the same manual action hundreds of times. A new technology makes something possible that was impossible several years earlier.
There is always another challenge waiting.
Automation as a Mindset
Automation became part of my thought process very early.
One of my first significant projects involved automatically collecting radiology reports, packaging them into a nightly archive, and transferring them to a billing company. The organization knew the reports needed to reach the billing company but did not have a defined technical solution for accomplishing it. So, I built one, and that process ultimately operated for approximately 15 years.
Another early project involved installing PACS viewing software for referring physicians. The installation required multiple manual steps: installing software, configuring a VPN client, importing the appropriate VPN profile, and modifying configuration values within an INI file.
After manually installing several computers, I went home and automated the process. By the next opportunity, much of the installation had become essentially one click. The motivation was not simply to avoid work. In fact, automation frequently requires considerably more work initially.
The reward is leverage.
If a computer can reliably perform a repetitive process, the person who previously performed that process is free to solve another problem. A few hours of engineering can sometimes eliminate hundreds of hours of future manual effort. Automation therefore is not necessarily about doing less.
It is about increasing the amount of useful work one person can accomplish.
That distinction has remained important throughout my career and has become even more significant as artificial intelligence expands what can reasonably be automated.
Autonomy and Experimentation
Technical curiosity alone, however, is not enough. It requires room to operate.
There were periods during my career when my ability to experiment was limited. I continued learning and solving problems, but organizational constraints restricted how aggressively I could change systems or explore new approaches. When those constraints eventually decreased and I was given substantially greater freedom to experiment, the difference in output was dramatic.
Within a relatively short period, ideas that previously might have remained ideas became working systems, integrations, automations, and applications. This experience has made me believe that autonomy may be an important but underappreciated factor in developing exceptional informatics professionals.
An organization can hire someone who is curious, technically capable, and motivated, then unintentionally train those characteristics out of them by discouraging experimentation.
“Don’t touch it.”
“That is the vendor’s responsibility.”
“Just open a ticket.”
“We have always done it this way.”
Those responses may sometimes be appropriate, particularly in a clinical environment where reliability and patient safety must take precedence over experimentation. Responsible experimentation requires controls, testing, and an understanding of consequences.
But making avoidance the default has a cost. People cannot develop deep understanding of systems they are never allowed to explore.
The opposite environment can have a multiplying effect. Give a curious person appropriate access, responsibility, accountability, and freedom to experiment, and the result may extend far beyond maintaining existing systems.
They begin creating new ones.
A Particular Kind of Mind
I cannot claim that every successful imaging informatics professional thinks exactly as I do, nor should success in the field require a single personality type. However, my experience has made me suspect that many of the people who become deeply passionate about this field share certain characteristics.
- They are curious.
- They enjoy difficult problems.
- They want to understand how systems work beneath the interface presented to users.
- They remember solutions and apply lessons from one system to another.
- They see inefficiency and instinctively begin considering how it could be eliminated.
- They are persistent enough to spend hours solving something that may ultimately save someone else only a few seconds.
- They take satisfaction in mastery.
And perhaps most importantly, they care about the result.
Imaging informatics continuously rewards these characteristics because the domain is broad enough that complete mastery is effectively impossible.
There is always another layer.
Meaning Behind the Technology
Challenge alone cannot completely explain the attachment. Imaging informatics also provides something that many technical disciplines cannot provide as visibly: a connection between the technology and something that matters.
The data moving through these systems represents real people.
The image that successfully arrives at a workstation is not merely a successful DICOM transaction. It may be the CT examination that identifies a patient’s stroke, the radiograph that confirms placement of a device, the mammogram that identifies a cancer, or the prior study that allows a radiologist to recognize an important change. None of this means the informatics professional interprets the exam or provides the medical care.
But we build and maintain part of the infrastructure that allows those things to happen.
When a complex imaging environment operates correctly 24 hours a day, 365 days a year, there is tremendous satisfaction in knowing that you helped make that possible.
There is even greater satisfaction when you helped design it.
Conclusion
I began my career knowing computers and knowing nothing about radiology.
Sixteen years later, I still encounter things I do not know.
That may be exactly why I am still here.
Imaging informatics offers an unusual combination of technical depth, clinical complexity, constant change, visible impact, and nearly unlimited opportunities for problem solving. For people whose personalities are driven by curiosity, experimentation, automation, and mastery, it can provide a nearly continuous cycle of challenge and reward.
That same curiosity and desire to solve problems eventually became part of the motivation behind building RadFuzion. It was not simply about creating another software platform. It was about applying years of experience in imaging informatics to the problems I had encountered firsthand, and building the tools I wished had existed each time I was the person responsible for making everything work.
The technology has changed dramatically during my career. On-premises PACS environments have expanded into interconnected ecosystems involving cloud infrastructure, enterprise imaging, increasingly sophisticated interoperability, advanced automation, and artificial intelligence. The problems I solve today would have been difficult to imagine when I started.
Yet the motivation to solve them feels remarkably similar:
- Find the problem.
- Understand it.
- Experiment.
- Solve it.
- Make the solution reliable.
- Make it easier for everyone who comes after you.
- Then find the next problem.
Perhaps that is why some people do more than simply work in imaging informatics.
They fall in love with it.