EMB Tech Talk Luncheon, "Protecting Patient Identity Integrity in AI-Enabled Healthcare"
The IEEE Lone Star Section (LSS)'s Engineering in Medicine and Biology (EMB) Chapter and Life Members Affinity Group (LMAG) is hosting the following Technical Talk Luncheon.
Registration Details: Open to the public -- you do not need to be an IEEE member to attend! There is limited seating with seating going to those who register via this vtools event in the order received; plan ahead to REGISTER!
Registration Fees: At the door, there is a $20.00 fee for IEEE EMB or LMAG Members, $30.00 for Guests, and $5.00 for Students. We accept Cash or Check, no credit cards. Exact amounts preferable. Remainder of the tab will be picked up by the EMB.
Date and Time
Location
Hosts
Registration
-
Add Event to Calendar
Speakers
Saiteja of CVS Health
Protecting Patient Identity Integrity in AI-Enabled Healthcare
Reliable patient identity matching is fundamental to healthcare data interoperability, clinical information integrity, and responsible AI adoption. When two patient records are incorrectly joined, downstream systems can associate information with the wrong patient, even when individual access controls function correctly. My work involves patient identity matching at national scale, across more than 185 million people. The challenging cases illustrate why identity resolution requires more than simple demographic matching. Twins may share a birth date and have similar names. Relatives can share addresses and phone numbers. A surname corrected after records were previously linked may require an accepted link to be reopened and reconciled. My proposed session focuses on a two-gate release approach for addressing both privacy and matching reliability. The first gate checks salt-hashed Bloom filters and fails the release if a frequency probe can recover common names. The encoding approach was presented at IEEE INTCEC. The second gate measures precision and recall against a fixed synthetic hard-case set. I hold a release until both gates pass. I will discuss why this validation matters for AI-enabled healthcare systems. An AI agent can pass its permission checks and still treat a wrong patient join as fact. Reliable identity relationships are therefore an important foundation for trustworthy clinical data processing and AI-assisted workflows. The presentation uses synthetic records exclusively, with no PHI, and emphasizes the technical methods and their implications for biomedical informatics and healthcare engineering.
Biography:
Saiteja Jonnalagadda is an IEEE Senior Member and a Senior Cloud Engineer at CVS Health. He is an applied-AI specialist who designs large-scale distributed systems where artificial intelligence has to work reliably in production. He holds a Master of Science in Applied Computer Science from Northwest Missouri State University and has spent close to seven years across healthcare and financial services building the kind of backend that runs continuously and cannot afford to be wrong, from event-driven microservices and real-time data pipelines to machine-learning components that make decisions on live data at scale. His core strengths are distributed-systems architecture, applied machine learning and generative AI, anomaly detection, data-quality engineering, and cloud-native platform design, built with Java, Spring Boot, Go, Apache Kafka, Google Cloud Platform, and Kubernetes. Saiteja is active in the technical community. He serves on the Technical Program Committees of several 2026 international conferences, including the IEEE World AI IoT Congress, has peer-reviewed more than fifteen papers across thirteen international conferences in artificial intelligence, machine learning, and cloud computing, and has judged student and academic technology showcases at institutions including Georgia Tech and Kennesaw State University. He is an active member of the IEEE.
Email:
Agenda
1100 - Arrival
1130 - Lunch and Presentation
1300 - Adjourn