Methods and systems for acute stress detection

WO2026127959A1PCT designated stage Publication Date: 2026-06-18HARMAN INT IND INC
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
HARMAN INT IND INC
Filing Date
2024-12-11
Publication Date
2026-06-18

AI Technical Summary

⚠Technical Problem

Existing automated stress detection systems face challenges in accurately identifying acute stress episodes due to wide variance in human responses and the difficulty in generating high-quality ground truth data for machine learning models, leading to poor accuracy and ineffective interventions.

⚗Method used

A method involving a stress detection system that uses a machine learning model trained to detect distinct stages of acute stress through inter-beat interval (IBI) data, characterized by a predefined threshold reduction and duration, utilizing a three-stage pattern recognition approach (activation, resistance, and recovery) to label cardiac data accurately.

🎯Benefits of technology

Enhances the accuracy of stress detection by generating high-quality training data, allowing the model to identify stress episodes more effectively and enabling timely interventions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure US2024059667_18062026_PF_FP_ABST
    Figure US2024059667_18062026_PF_FP_ABST
Patent Text Reader

Abstract

Methods and systems are described for detecting acute stress in a human subject based on heart rate (HR) data acquired from the subject, using a machine learning (ML) model. The model may be trained and / or validated on inter-beat interval (IBI) data extracted from physiological signals (ECG, PPG, sonogram, echogram, etc.), using labeled IBI data as ground truth data. In particular, the labeled IBI data labels three stages of an episode of stress: an activation stage, a resistance stage, and a recovery stage. By labeling the three stages, a performance of the model during training may be increased with respect to other approaches to stress detection. The labeled IBI data may also be used as ground truth data for different input datasets not including HR data.
Need to check novelty before this filing date? Find Prior Art