Episode
How AI Is Spotting Sepsis Hours Before Nurses Can
- Published
- Jun 17, 2026
- Duration seconds
- 696
- Processing state
not_requested
Actions
POST https://stenobird.com/v1/public/podcasts/the-digital-health-podcast-with-fexingo-wearables-diagnostics-and-health-software-7872006/episodes/how-ai-is-spotting-sepsis-hours-before-nurses-can/transcription-requests
Idempotently request low-priority transcript generation for this episode.GET https://stenobird.com/podcast/the-digital-health-podcast-with-fexingo-wearables-diagnostics-and-health-software-7872006/how-ai-is-spotting-sepsis-hours-before-nurses-can.md
Read the agent-friendly Markdown representation of this episode resource.
Summary
Sepsis kills 11 million people a year globally, often because early symptoms look like the flu. But a new wave of hospital AI systems—trained on continuous vital-sign streams from wearable sensors—is catching it three to six hours before traditional screening tools. In this episode, we dive into the specific algorithms running at Johns Hopkins and the UK's NHS, how they combine heart rate variability, respiratory rate, and temperature trends into a single risk score, and why doctors still hesitate to trust a machine's warning over their own instincts. We also explore the tension between faster detection and alert fatigue—a real problem when the AI flags 40 percent of patients as at risk. No hype, just the numbers: false positive rates, time saved, and what it means for the future of hospital monitoring. #SepsisDetection #AIinHealthcare #WearableSensors #HospitalAI #HealthTech #MachineLearning #JohnsHopkins #NHS #VitalSigns #PredictiveAnalytics #PatientSafety #ClinicalAI #AlertFatigue #HealthInnovation #DigitalHealth #Business #FexingoBusiness #BusinessPodcast Keep every episode free: buymeacoffee.com/fexingo