Episode
How AI Is Predicting Sepsis Hours Before It Strikes
- Published
- Jul 7, 2026
- Duration seconds
- 528
- Processing state
not_requested
Actions
POST https://stenobird.com/v1/public/podcasts/healthtech-talks-with-fexingo-digital-health-telemedicine-and-medical-software-7871821/episodes/how-ai-is-predicting-sepsis-hours-before-it-strikes/transcription-requests
Idempotently request low-priority transcript generation for this episode.GET https://stenobird.com/podcast/healthtech-talks-with-fexingo-digital-health-telemedicine-and-medical-software-7871821/how-ai-is-predicting-sepsis-hours-before-it-strikes.md
Read the agent-friendly Markdown representation of this episode resource.
Summary
Sepsis kills roughly 350,000 adults in US hospitals each year — often because clinicians catch it too late. In this episode, Lucas and Luna drill into a single case: how the University of Florida Health system deployed a real-time AI sepsis alert that flags patients 4 to 6 hours before they meet standard diagnostic criteria. They discuss the model's design — it analyzes 35 vitals and lab values every 5 minutes — and the hard part: getting doctors and nurses to trust and act on the predictions. Lucas shares a specific stat from a 2024 study: the algorithm reduced sepsis mortality by nearly 19 percent in the intervention group. Luna pushes back on alarm fatigue and false-positive rates. A focused episode on one of the most promising — and trickiest — applications of machine learning in critical care. #SepsisPrediction #ClinicalAI #PatientSafety #RealTimeMonitoring #CriticalCare #UFHealth #MachineLearning #HealthcareAI #MedicalSoftware #DigitalHealth #HospitalTech #PredictiveAnalytics #Business #Technology #HealthtechTalks #FexingoBusiness #BusinessPodcast #PedroFonseca Keep every episode free: buymeacoffee.com/fexingo