Emotional event recognition method and system based on multi-modal physiological parameters

By analyzing multimodal physiological parameters and utilizing the synergistic changes in respiratory rate, heart rate, and blood pressure, an emotional event recognition method was constructed. This method addresses the problem of neglecting the response speed and intrinsic regulatory connections of physiological parameters, achieving higher recognition accuracy and individual adaptability.

CN122272022APending Publication Date: 2026-06-26FIRST AFFILIATED HOSPITAL OF KUNMING MEDICAL UNIV
0 Cites 0 Cited by

Patent Information

Application Number
CN202610394278.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-03-28
Publication Date
2026-06-26

AI Technical Summary

Technical Problem

Existing technologies neglect the response speed and intrinsic regulatory connections of physiological state changes in different physiological parameters, resulting in insufficient accuracy in identifying emotional events and difficulty in adapting to the physiological regulatory patterns of different individuals.

Method used

By collecting multimodal physiological parameter data in real time, a time series of physiological parameters is generated. A baseline time window is determined based on the respiratory rate time series. The analysis interval is divided according to the hysteresis response characteristics of heart rate and blood pressure. A respiratory-heart rate correlation model is constructed. Blood pressure changes are analyzed by combining heart rate variability parameters. A set of historical characterization features is introduced for weighted fusion optimization, and emotional event category labels are output.

Benefits of technology

It improves the accuracy and individual adaptability of emotional event recognition, breaks through the limitations of single-parameter analysis, and enhances the richness of feature dimensions and recognition accuracy.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention relates to the technical field of emotion recognition, in particular to an emotion event recognition method and system based on multi-modal physiological parameters. The method comprises the following steps: determining a reference time window according to a change trend of a respiratory frequency sequential sequence; respectively determining a heart rate analysis interval and a blood pressure analysis interval according to respective lag response characteristics of the heart rate sequential sequence and the blood pressure sequential sequence in the reference time window; current characterization features are formed by the reference time window, the heart rate and the blood pressure analysis interval; determining a feature matching degree according to the similarity between the current characterization feature and a preset historical characterization feature set; according to the feature matching degree, performing weighted fusion optimization on the current characterization feature, and determining an optimized characterization feature; and inputting the optimized characterization features into a pre-training classification model, and outputting category labels corresponding to the emotional events, so that the characterization features better fit a real emotional-physiology linkage rule, and a classification result can give consideration to the difference of individual physiological features.
Need to check novelty before this filing date? Find Prior Art