Some people get a job offer based on decades of relevant industry experience. Others land the dream position thanks to their academic credentials. And for some, the door opens because of whom you know. For software engineer Parham Gholami’s current employment, it was none of the above.
“I wanted to make it easier for people with limited computer experience to visualize and interact with the complex three-dimensional datasets that any radiology lab produces.”
That’s according to Gholami’s hiring manager and current boss, Justin Ryan, PhD, a research scientist and director of the Helen and Will Webster Foundation 3D Innovations (3DI) Lab at Rady Children’s Hospital – San Diego (RCHSD). It is the largest such facility on the West Coast, and the only one in the area dedicated exclusively to pediatric health and trauma care.
Needless to say, it’s important work. As discussed in a Voices AMplified article from April 2023, Ryan has spent the last five years building a lab that converts images taken via computed tomography (CT) scanning and magnetic resonance imaging (MRI) into digital models, then 3D prints highly accurate replicas for surgical planning, training of medical staff, evaluation of tools and techniques, and to help alleviate the angst that young patients and their families might feel about an upcoming procedure.
It’s ironic, then, that while Gholami has extensive experience with C++, Python, Java and other computer programming languages—not to mention his current pursuit of a master’s degree in computer science from the Georgia Institute of Technology—he knew virtually nothing about additive manufacturing (AM) when he accepted the job at Rady.
That’s all changed over the last few years. Gholami is now an integral part of a team that manages an assortment of high-end 3D printers. These include an HP Jet Fusion 580 Color 3D printer, a Stratasys J750 Digital Anatomy printer and a F370 FDM machine from the same manufacturer, as well as several stereolithography machines from the likes of 3D Systems and Formlabs. While Gholami occasionally operates these printers or post-processes the parts that come out of them, his focus is on something more foundational: The 3D models and 3D-printed parts the lab produces would be less effective or even impossible to develop without his continuing efforts.
That’s because he has created a software program with an intuitive user interface that allows surgeons and other medical practitioners to more easily review and analyze the digital models generated from the hospital’s immense CT and MRI datasets. “We built a 3D viewer that greatly simplifies the process compared to the legacy radiology tools, which had dozens of buttons that a surgeon would never use,” he notes. “It’s far more intuitive.”
Without rehashing the technical details described in the earlier article, it’s important to note that Gholami also developed the means to interface with and enhance the hospital’s legacy software systems, such as its PACS (Picture Archiving and Communication System) database. The result is Media2DICOM, a downloadable tool available on the RCHSD website that allows users to encapsulate various media—including 3D-printable files—in the industry standard DICOM (Digital Imaging and Communications in Medicine) format.
As Ryan said in that earlier interview, Media2DICOM “supports the addition of metadata, which is not only a huge time saver but eliminates the possibility of file naming errors and other potential problems.”
Another irony is the manner in which he and Gholami came to know one another. There were no recruiters involved. No job postings on SimplyHired or Indeed. Instead, Ryan advertised the job opening on an unorthodox yet—in hindsight—brilliant medium: Reddit.
“There’s a community there for people who follow Unity, a video game engine that I’ve used both as a hobbyist and in my professional life,” Gholami confides. “Anyway, I was browsing there one day and saw a post about working at a local children’s hospital in San Diego. The projects it described sounded incredibly interesting, so I replied to the person who posted it and asked for additional details. That person was Justin Ryan.”
Programming credentials to the contrary, Gholami says he was “incredibly fortunate” to get the job. RCHSD hired him as a research associate a few months before the global pandemic, and he was promoted to research engineer soon after.
Since then, Gholami has developed the software interface tools described, as well as some that are less sexy but equally important. “In a lot of ways, my work here is about solving problems for the hospital,” he says. “Many are related to improving patient care and making life easier for the doctors and researchers, but others are more utilitarian.”
One example of this evolved during the early days of COVID-19, when hospital administration struggled with the possibility of having too many patients and not enough beds. Gholami developed Delta Planner, a room planning application that addressed fears over potential patient surges and allowed staff to determine the best way to safely fit everyone—whether ill with the virus or not—into the hospital.
“It wasn’t all that different from developing a level editor for a game, like something you would find in The Sims or Animal Crossing,” he says. “You have a virtual space to work with, which in this case was the floor plan for the hospital’s Acute Care Pavilion, and the application allows you to measure the distances between beds, stretchers, cribs and different types of equipment, and place them in the most optimal positions possible.”
Fortunately, RCHSD never had to use Delta Planner for its original purpose. But Gholami soon had an opportunity to revisit the code. While evaluating initial construction plans for the Dickinson Image-Guided Intervention Center, an innovative pediatric unit that “houses novel real-time magnetic resonance imaging technology” and minimizes exposures to harmful radiation, some of the staff were concerned about the overall flow of moving patients into and out of some of the rooms.
“I modified Delta Planner so that we could take a 3D-digital reconstruction of the center and move virtual beds around, identifying pinch points and other possible trouble spots. It was pretty cool, because we were able to do the entire thing on a massive Zoom call so everyone concerned could weigh in.”
Less cool but still important was Gholami’s third iteration of Delta Planner, where he used the tool to assist the hospital’s logistics team with the relocation and subsequent reconstruction of its loading dock. Here, he used blueprints of the proposed space to create a 3D model and then populated it with virtual delivery vans and semi-trucks to analyze traffic flow. “That’s a great example of custom software with a lot of flexibility, since it allows different parts of the hospital to use it outside of what we consider traditional patient care. Either way, it’s about helping to solve whatever challenges come my way.”
Gholami is currently revisiting another of his legacy tools, the 3D viewer that Ryan tasked him with building early in his employment. Here again, the goal is to simplify working life for medical professionals, who often lack the knowledge, training or time needed to navigate complex software applications. And one of the ways Gholami and Ryan aim to accomplish this is with virtual reality (VR) and augmented reality (AR) technologies that will one day allow a surgeon to “fly through” a virtual heart, for example, or pick up a computer-generated scalpel and practice a procedure in a simulated operating room.
As with the day when Gholami was surfing Reddit, his experience with Unity plays a significant role in these efforts. When writing the original viewer, he leveraged the well-known gaming engine for its graphics rendering capabilities. This not only made the software both easy to use and visually accurate, but has eased the development of subsequent revisions.
“We’re looking at AR/VR for a number of uses,” he says. “Just as a clinician might don a headset and peer inside a patient’s virtual organs, the people who will be working on the loading dock could tour the area while it’s still in the planning stages.
“We’re also evaluating the possibility of co-pilots, who will sit outside the simulation and make adjustments based on the AR/VR user’s feedback,” Gholami continues. “Regardless, the technology gives you a physical sense of how things are laid out, the area’s scale and functionality, and in the case of surgical planning, will lead to better patient outcomes. Ultimately, that’s what my job here is all about.”