Personalized baseline correction for detection of infection from physiological signals
Patent Information
- Authority / Receiving Office
- EP · EP
- Patent Type
- Applications
- Current Assignee / Owner
- KONINKLIJKE PHILIPS NV
- Filing Date
- 2024-07-12
- Publication Date
- 2026-05-27
AI Technical Summary
Existing unobtrusive vital sign monitoring systems face challenges in detecting early onset symptoms of infections due to individual variability in physiological signals, which complicates the training and accuracy of AI models.
A personalized baseline correction system that uses a calibration process to create a personalized normalization process for each patient, considering the entire distribution of baseline values rather than just the mean and variance, to standardize physiological data and improve detection of subtle vital sign changes.
The system effectively reduces data variability, improves normalization of patient monitoring data without assuming a Gaussian distribution, minimizes data requirements for AI models, and enhances the detection of early onset symptoms of infections.
Smart Images

Figure EP2024069872_23012025_PF_FP_ABST