Behind the Software: Flywheel Engineer Profiles – Pablo Velasco, Senior Scientific Solutions Engineer
Quick facts:
-
Former Chief MRI Physicist at the Center for Brain Imaging of New York University
-
Spent 17 years as a research scientist and senior research scientist and two years as a postdoctoral scholar
-
Originally from Spain and now lives in Southwest Florida
Pablo Velasco, Senior Scientific Solutions Engineer for Flywheel, knows his way around the imaging research world. Having spent nearly two decades in academia, in roles overseeing all things data science and MRI imaging, his journey has now led him to Flywheel. Over the past two-and-a-half years as a Senior Scientific Solutions Engineer, he’s dedicated himself to taking his background in imaging research and passion for coding and applying it to helping clients solve their greatest challenges around imaging research.
A scientist’s journey to Flywheel
Pablo’s professional life began in Madrid, where he pursued a PhD in physics. He later came to the U.S. as a postdoctoral scholar, followed by roles as a research scientist and a senior research scientist at NYU, overseeing operations, development of pre-processing tools, automated reproducible workflows and more in the MRI space for nearly 20 years.
“I thought MRI was fascinating. It was amazing to see how these images could uncover the brain’s inner workings and function,” Pablo shares.
After a long career in academic research, Pablo decided it was time for a change. In the latter part of his academic work, he had been working on developing data processing pipelines and discovered the exciting applications of new technology to this process, which naturally led him to Flywheel.
Making a difference at Flywheel
As a Senior Solutions Engineer at Flywheel, his role includes a range of responsibilities — from helping train and consult with clients on their Flywheel instance to developing custom workflows and gears. Each day, Pablo finds himself working directly with clients to address their challenges, offer support and build solutions.
“In general, I spend most of my time meeting with customers and helping with service tickets,” Pablo says. “But I also love developing new solutions and features that help integrate their platforms, automate workflows and create reports.”
For Pablo, the most rewarding aspect of his work is helping clients overcome research roadblocks. In fact, he said he’s found from his experience in academia that the oft-cited statistic that 80 percent of researchers’ time is spent preparing data rather than analyzing it holds true.
“Flywheel allows researchers to focus on what really matters by automating data curation and organization, so they can concentrate on the final analysis of their data,” he says.
Tackling challenges in data harmonization
Pablo notes that today’s researchers face significant challenges, primarily around data harmonization and scalability. Many clients come to Flywheel because they’re grappling with messy, inconsistent data from multiple sources or struggling to find the computing power to process large datasets.
“Researchers often have to combine data from different sources, which leads to mismatched file names, missing information and variable data quality,” Pablo explains.
Flywheel’s cloud-based infrastructure offers a scalable solution to these challenges, allowing researchers to access the computing power they need without maintaining costly on-premises systems. Pablo walks us through one account of a Flywheel client with variable computing needs and how the platform is able to scale up and down computing as needed: “One of our customers came to us with several thousand jobs that were taking weeks to process. They needed the data processed faster, so we proposed scaling up their cluster. Instead of using eight machines, we increased it to 64 and upgraded the processing power of each one. The jobs were completed in a matter of days rather than weeks. The cost of computing is similar whether you scale up temporarily or maintain a small, constant footprint. And because it’s all cloud-based, there’s no need to buy physical machines.”
Pablo notes that, in the world of academia, researchers often have to maintain their own servers, swap out disks when they break and deal with outdated machines. That process can be time-consuming, costly and tedious, and within a few years, the hardware may be obsolete.
“At Flywheel, our customers don’t have to worry about any of that,” he says. “They can access the computing power they need almost instantly, and when they’re done, those resources are released, and they’re no longer charged. This setup reduces hardware and personnel costs, allowing faster data analysis.”
The power of Gears: automating data workflows
One of the standout features of Flywheel’s platform is its use of Gears—customizable algorithms that automate tedious jobs. Pablo says he loves getting into the nitty gritty of Gears with clients and notes that custom Gears are game changers because they let researchers automate workflows, scale their processing and ensure they’re always using the same settings.
He highlights the “hierarchy-curator” Gear as a favorite: “This Gear lets you run Python scripts on your data, which is excellent for prototyping. You can make small changes in the script to see how the results change without the need to build a new Gear and upload it to the platform. For a developer, that’s really useful.”
As far as the most valuable Gear to customers, that has to be the “file-metadata-importer,” Pablo says. This Gear extracts crucial information from the DICOM header, such as the patient code, image type, parameters, resolution and the instrument used.
“Typically, the only way to access this information is to download the file and examine it directly,” he says. “This tool, however, pulls all that data and attaches it to the database.”
Outside of work: a love for nature and family
For Pablo, it’s not all Gears and coding—he lives in sunny southwest Florida and enjoys swimming, spending time in prayer and going on long walks with his wife. Even when he’s working, he finds ways to unwind: “I do a lot of coding on my porch, listening to the birds and feeling the breeze.”
Flywheel’s medical imaging and AI experts are ready to help with your individual imaging data needs. Whether it’s developing algorithms, testing them or helping you gain visibility into all of your files, we can help you get more from your imaging data. Schedule a call to get started.
