Wearable Optical Skin Tone Estimation for Stable Physiological Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wearable devices face inaccuracies in physiological data collection due to varying skin tones, which affect light absorption and signal quality, leading to inconsistent performance of automatic gain control components.
Innovation Solution
Implementing a machine learning model to estimate skin tone by analyzing signal strength and transmit power parameters, allowing for validation of component functionality and adjustment of measurement parameters based on skin tone.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If automatic gain control components are used to maintain signal strength within a range, then signal quality is improved, but performance consistency across different skin tones deteriorates
Solution Approach 1:
The patent applies parameter changes by using a machine learning model to estimate skin tone metrics from optical signal characteristics (signal strength and transmit power). Based on the estimated skin tone, the system dynamically adjusts measurement parameters such as optical power levels and signal processing gains to compensate for varying light absorption across different skin tones, thereby maintaining both signal quality and performance consistency.
2Productivity
If light absorption characteristics are used for physiological data collection, then physiological data acquisition is enabled, but accuracy across different skin tones deteriorates
Solution Approach 1:
The patent implements feedback by using the estimated skin tone metric to adjust subsequent measurements. The machine learning model continuously estimates skin tone from optical signals, and this estimation feeds back into the measurement system to modify transmit power and signal processing parameters, creating a closed-loop system that maintains accuracy across different skin tones while enabling continuous physiological data acquisition.
Solution Approach 2:
The system changes measurement parameters (optical power, gain settings) based on the estimated skin tone to compensate for differential light absorption. This allows the system to maintain accurate physiological measurements across diverse skin tones while preserving continuous data acquisition capability.
3Measurement precision
If skin tone estimation is added to compensate for variations, then accuracy is improved, but device complexity increases
Solution Approach 1:
The patent introduces an intermediary machine learning model that estimates skin tone from existing optical signal measurements. This intermediary component translates raw signal data into skin tone metrics without requiring additional sensors or hardware, thereby improving accuracy while minimizing the increase in device complexity by reusing existing optical components.
Solution Approach 2:
The system uses its own existing optical signals (measured for physiological data collection) to estimate skin tone, rather than requiring separate measurement systems. This self-service approach allows the device to compensate for skin tone variations using data already being collected, avoiding the need for additional hardware 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
Enhances the accuracy of physiological data collection by compensating for skin tone variations, ensuring consistent performance of wearable device components across different skin tones.
Implementation Method 1
an optical transmitter and an optical receiver. The optical transmitter may be configured to transmit light associated with a first wavelength using a first set of one or more light-emitting components
Implementation Method 2
a skin tone of the user may impact the signal quality of the physiological data. For example, a user with a darker skin tone may absorb a higher proportion of light than a user with a lighter skin tone
Data Source
AI summary
Methods, systems, and devices for estimating skin tone using a machine learning (ML) model are described. A wearable device may transmit light using a light-emitting component, and generate a signal based on the light received at a photodetector. In some examples, a strength of the received signal may be maintained within a signal strength band. The wearable device may identify a data set pair based on the signal, the data set pair including the signal strength band and a transmit power parameter that corresponds to the signal strength band. The data set pair may be inputted into an ML model and the ML model may output a skin tone metric for the user. The estimated skin tone metric may then be used to adjust measurement parameters used by the wearable device to improve a quality of physiological data, or validate algorithms used by the wearable device across skin tones.