Straight Outta Health IT
Healthcare AI can only be as strong as the data behind it, but patient privacy cannot be treated as collateral damage. In this episode of Straight Outta Health IT, Dr. Julia Komissarchik, CEO and Founder of Glendor, Inc., joins Christopher Kunney to explore one of the most pressing challenges in healthcare AI today: how to responsibly use patient data while ensuring that protected health information (PHI) is not exposed. Julia breaks down why de-identification is far more complex than simply stripping a patient’s name from a record. She explains that PHI can be embedded in many different forms beyond structured fields, including clinical text, medical images, voice recordings, video, metadata, and even indirect contextual signals such as hospital names, locations, or physician specialties. Because of this, ensuring true privacy protection requires a much more nuanced and multi-layered approach than many assume. She also highlights the importance of diversity in healthcare data when building AI systems. According to Julia, robust models depend on exposure to a wide range of populations and care settings, including rural, urban, tribal, and underserved communities, to ensure that AI performs reliably and equitably across different real-world environments. Tune in to learn why privacy, trust, and responsible data infrastructure are foundational to the future of AI in healthcare. Resources * Connect with Dr. Julia Komissarchik on LinkedIn here [https://www.linkedin.com/in/juliakomissarchik/] or reach out via email here [julia@glendor.com]. * Follow Glendor, Inc on LinkedIn here [https://www.linkedin.com/company/glendor/] and visit their website here. [https://glendor.com/]
106 afleveringen
Reacties
0Wees de eerste die een reactie plaatst
Meld je nu aan en word lid van de Straight Outta Health IT community!