Biometric Stress Calculation Using Segmented Feature Aggregation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for calculating stress values and physical conditions of individuals using mid- to long-term biometric data face challenges in processing large data sets, leading to prolonged calculation times and difficulties in obtaining prompt results.
Innovation Solution
A calculation method that involves acquiring time-series biometric data, calculating minimum feature values for preset time units, and then using these values to compute feature values for longer time units, ultimately determining a value representing the physical condition of an individual.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If mid- to long-term biometric data is processed together to calculate stress values, then calculation accuracy is improved, but calculation time increases significantly
Solution Approach 1:
The patent divides long-term biometric data into multiple preset time units (e.g., daily, weekly, monthly periods). For each time unit, it calculates minimum feature values (such as minimum, maximum, average heart rate) and then aggregates these pre-calculated values to derive the final stress indicator. This segmentation approach reduces computational complexity by avoiding processing of all raw data points simultaneously, thereby maintaining accuracy while significantly reducing calculation time.
2Reliability
If large amounts of entire biometric data are processed, then comprehensive physical condition assessment is achieved, but processing efficiency decreases
Solution Approach 1:
The patent performs preliminary calculations of minimum feature values for each preset time unit in advance. These pre-calculated feature values (minimum heart rate, maximum heart rate, average heart rate for each period) are stored and reused when calculating stress indicators. This preliminary action eliminates the need to re-process entire raw biometric datasets repeatedly, ensuring reliable comprehensive assessment while dramatically improving processing efficiency.
3Measurement precision
If detailed biometric data processing is performed, then accurate stress indication is obtained, but real-time calculation capability is compromised
Solution Approach 1:
The patent segments biometric data into discrete preset time units and calculates key feature values (minimum, maximum, average) for each segment. When a new time unit is completed, the system immediately calculates the stress indicator using the pre-computed feature values from all time units. This segmentation enables the system to maintain accurate stress indication while achieving real-time calculation capability, as the computational burden is distributed across manageable segments rather than requiring processing of all data simultaneously.
Data Source
AI summary
A stress value calculating device 100 of the present invention includes: a minimum feature value calculating unit 121 that acquires biometric data obtained by measurement in a time series from a person, and calculates, as a minimum feature value, a feature value of each preset minimum time unit of the acquired biometric data; a first feature value calculating unit 122 that calculates, after passage of a first time unit which is a time unit longer than the minimum time unit, as a first feature value, a feature value of biometric data obtained by measurement in the first time unit using the minimum feature value corresponding to the biometric data of the first time unit; and a calculating unit 123 that calculates a value representing the physical condition of the person using information based on the first feature value.