We’re excited to be back for another episode of the Life Sciences Today Podcast by Healthcare IT Today. My guest today is Asta Ratkeviciene, Co‑Founder and CEO at MedSyntra. In this episode of Life Sciences Today, I talk with Ratkeviciene about building a global, ethically sourced medical imaging data infrastructure across radiology, pathology, oncology, and women’s healthcare. Ratkeviciene explains how MedSyntra aggregates and de-identifies millions of imaging studies, provides a highly curated data marketplace and platform for institutions that require data sovereignty, and partners with clinical experts for annotation and external teams for model development. They discuss why imaging is the core focus, how population differences can degrade AI performance across geographies, MedSyntra’s evolving business model, and the biggest anti‑patterns Ratkeviciene sees in the medical AI ecosystem today.
Check out the main topics of discussion for this episode of the Life Sciences Today podcast:
- What’s the most important thing people misunderstand about medical imaging data?
- Who must have this infrastructure before they can deploy serious AI?
- When did you realize you weren’t building a dataset—but infrastructure?
- What early decision optimized for long-term trust rather than speed?
- Where does ethical data sourcing actually slow you down commercially?
- Have you walked away from data that would have helped the business—but violated your principles?
- Do you think ethical language is sometimes used to hide poor data practices?
- Why do imaging AI models quietly fail when they cross borders?
- When does population bias show up—training or deployment?
- Why can’t this be solved with more data alone?
- How do you reconcile local data control with global AI learning?
- Do we end up with federated intelligence—or fragmented ecosystems?
- What’s harder: collecting imaging data or producing clinically meaningful labels?
- If annotation quality doubled, what would that unlock that most people aren’t expecting?
- What’s the biggest anti-pattern you see in medical AI today?
- Ten years from now, what will seem obvious about medical imaging AI that people are still ignoring today?
Subscribe to Danny’s newsletter to get strategic patterns for life science leaders building a defensible business.
Be sure to subscribe to the Life Sciences Today Podcast on your favorite podcasting platform:
Along with the popular podcasting platforms above, you can Subscribe to Healthcare IT Today on YouTube. Plus, all of the audio and video versions will be made available to stream on Healthcare IT Today. As a former pharma-tech founder who bootstrapped to exit, I now help TechBio and digital health CEOs grow revenue—by solving the tech, team, and go-to-market problems that stall your progress. If you want a warrior by your side, connect with me on LinkedIn.
If you work in Life Sciences IT, we’d love to hear where you agree and/or disagree with our takes on health IT innovation in life sciences. Feel free to share your thoughts and perspectives in the comments of this post, in the YouTube comments, or privately on our Contact Us page. Let us know what you think of the podcast and if you have any ideas for future episodes.
Thanks so much for listening!
POWERED BY BLUBRRY
