The exploit of artificial intelligence tools in healthcare is accelerating as confidence in the results generated and delivered by such tools increases.
RAMP, one such tool already in operation at UVA Health, focuses on delivering actionable, verifiable and explainable machine learning, integrating it as a decision support tool into clinical workflows to improve insight into patient health trends and facilitate faster delivery of necessary care. care, improving patient outcomes.
AI-powered predictive analytics models leverage elaborate real-time and historical patient data to provide healthcare professionals with actionable information and alert care teams if a patient needs immediate facilitate.
Valentina Baljak is a senior data scientist at UVA Health. He is a PhD in computer science and technology, applied machine learning. UVA Health created and uses RAMP today.
Baljak and two of her colleagues will discuss artificial intelligence, RAMP and more at HIMSS25 in March in Las Vegas in a session titled “Real-time Analytics Monitoring Platform: Useful Artificial Intelligence in Action.” We spoke with Baljak to understand what she and her colleagues will be talking about during the session and what HIMSS25 attendees can take away from their talk.
Q. What is the main topic you will cover in your session and why is it relevant to healthcare and healthcare IT today?
AND. With the recent emergence of generative artificial intelligence models, this topic is gaining increasing attention in the healthcare field. In this work, we focus on real-time clinical decision support tools. Artificial intelligence is not a modern term.
At UVA Health, we’ve been developing real-time predictive systems for several years, and one of the biggest lessons we’ve learned is that the shape your AI should take best suits your needs. Doctors won’t exploit tools they can’t explain. Building trust in our models and tools meant working closely every step of the way, from day one.
We want to provide a blueprint for building a system that will work in your environment, and raise awareness of the importance of transparency, accountability and explainability of your models. This is especially critical in the medical environment, where real-time predictions can have a significant impact on patient outcomes.
Q. Your main focus will be on artificial intelligence. How is it used in healthcare in the context of the session topics?
AND. A key aspect of RAMP is real-time data collection from EHRs and other data sources. The ability to record results in the patient record in the EHR and notify care teams in real time makes RAMP a key tool in the clinical setting.
The technologies used here are quite established and all open source. Python provides a solid foundation for our development of machine learning, backend connectivity, and data processing. Connections to various data sources are built using FiHR, REST API and custom HL7. The website is built in Angular.
As part of our latest major expansion, we are building a modern predictive model based on our largest real-time data stream, built using Kafka to collect all vital signs and ECG waveforms from bedside monitors.
Q. Participants will come to your session wanting to take the knowledge home. What one takeaway can they expect?
AND. Artificial intelligence is a fundamental part of up-to-date healthcare and takes many forms depending on needs. Choosing the right approach to AI is critical given the high stakes.
If you have the in-house expertise and resources, developing a custom AI system is a powerful alternative to vendor-supplied black box systems.