AI Thermal Skin Temperature Normalization for Fever Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Temperature screenings using thermal sensors often inaccurately measure body temperature due to fluctuations in skin temperature caused by ambient temperature changes, potentially allowing individuals with fevers to enter environments undetected.
Innovation Solution
An AI system that normalizes thermal skin temperature to body temperature using neural networks, baseline management, and clustering techniques, accounting for differences between external and indoor ambient temperatures, and dynamically updates baselines to ensure accurate fever detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If thermal sensors are used for temperature screening, then non-contact measurement is achieved, but measurement precision deteriorates due to skin temperature fluctuations
Solution Approach 1:
The patent introduces an intermediary conversion model that maps skin temperature readings to body temperature estimates. This model acts as a mediator between the thermal sensor measurement (skin temperature) and the desired target (body temperature), using the relationship T_body = f(T_skin, T_ambient) to bridge the gap and provide accurate fever detection without direct body contact
Solution Approach 2:
The patent changes the measurement parameter from direct body temperature to skin temperature, which can be measured non-contactly. By adjusting and normalizing skin temperature readings based on ambient temperature conditions and conversion models, the system achieves accurate body temperature assessment indirectly through parameter transformation
2Productivity
If skin temperature is measured instead of body temperature, then measurement speed is improved, but reliability deteriorates due to ambient temperature influence
Solution Approach 1:
The patent implements feedback mechanisms that continuously monitor ambient temperature and use it to adjust the skin-to-body temperature conversion model in real-time. This feedback loop ensures that the conversion model remains accurate under varying environmental conditions, maintaining reliable fever detection while preserving fast screening speeds
Solution Approach 2:
The patent employs dynamic conversion models that adapt to changing ambient temperature conditions rather than using static relationships. The system dynamically adjusts the mapping between skin and body temperature based on current environmental parameters, ensuring continuous accuracy as conditions change throughout the day
3Measurement precision
If ambient temperature compensation is implemented, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent replaces complex physical measurement systems with computational models. Instead of using intricate hardware to directly measure body temperature, the system uses software-based conversion models that process thermal sensor data and ambient temperature readings through mathematical relationships, achieving high precision through information processing rather than mechanical complexity
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The AI system provides reliable fever detection by accurately converting skin temperature readings to body temperature, reducing false negatives and false positives, and effectively identifying individuals with fevers, thereby preventing the spread of contagious diseases.
Implementation Method 1
The temperature screenings may be performed by a thermal sensor
Data Source
AI summary
Methods, systems, and devices for data processing and artificial intelligence (AI) techniques are described. A system may support person detection using an optical camera and temperature detection using a thermal sensor. The thermal sensor may determine a temperature reading corresponding to a tracker for a specific detected person. A processing device may update a current temperature for the tracker based on the temperature reading and may add the updated current temperature to a baseline. For example, the processing device may use AI techniques to manage one or more baselines supporting handling of differences and changes in ambient temperatures. The processing device may determine that the current temperature for the tracker satisfies an alert threshold and may trigger an alert procedure. For example, the processing device may use an active baseline to correct the current temperature from a skin temperature to a body temperature for comparison to a fever detection threshold.