Voice Biomarker Screening for Blood Glucose With Feature Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing voice signal analysis systems face challenges in efficiently determining blood glucose levels in healthy individuals and as a potential biomarker for diabetes, requiring advanced methods to process voice data in real-time across various platforms and deliver accurate predictions.
Innovation Solution
A system and method using voice biomarkers selected based on longitudinal stability and decision-making ability, generating predictive models with a multi-class random forest classifier to determine blood glucose levels with high accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If voice signal analysis is performed to determine blood glucose levels, then non-invasive monitoring capability is improved, but processing overhead and computational complexity increase
Solution Approach 1:
The patent extracts and selects only the most relevant voice biomarker features from the complete voice signal spectrum. By identifying and isolating specific acoustic features that correlate with blood glucose levels, the system reduces the dimensionality of processing requirements while maintaining diagnostic accuracy, thereby resolving the contradiction between non-invasive monitoring capability and processing overhead
Solution Approach 2:
The patent applies local quality by focusing computational resources on specific, localized features within the voice signal that have been identified as biomarkers for blood glucose monitoring. Rather than processing the entire voice signal uniformly, the system concentrates analysis on particular acoustic characteristics that carry diagnostic information, optimizing the balance between monitoring reliability and processing complexity
2Measurement precision
If comprehensive voice signal processing is performed to extract all biomarker signals, then measurement accuracy improves, but processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary action by pre-identifying and selecting the most discriminative voice biomarker features before actual blood glucose measurement and prediction. This feature selection process, which includes evaluating feature importance and filtering out redundant features, is conducted in advance to create an optimized set of biomarkers that can be processed quickly during real-time monitoring, thus resolving the contradiction between measurement precision and processing time
Solution Approach 2:
The patent applies partial action by processing only the essential subset of voice features that are most strongly correlated with blood glucose levels, rather than performing exhaustive analysis of all possible voice signal characteristics. This selective processing approach maintains adequate measurement precision while significantly reducing computational burden and processing time
3Reliability
If multiple voice features are analyzed to improve prediction accuracy, then diagnostic reliability improves, but system complexity and processing overhead increase
Solution Approach 1:
The patent extracts and retains only the most informative voice features for blood glucose prediction by applying feature selection techniques. This process identifies and removes redundant or less discriminative features, maintaining diagnostic reliability through the retention of key biomarkers while reducing system complexity by eliminating unnecessary processing components
Solution Approach 2:
The patent applies universality by developing a multi-class random forest classifier that can simultaneously handle multiple voice biomarker features and predict different blood glucose categories. This unified modeling approach allows the system to process multiple features efficiently in a single computational framework, improving prediction accuracy without proportionally increasing system complexity
Data Source
AI summary
Provided are devices and systems for determining a blood glucose level for a subject, the device comprising a processor and a memory in communication with the processor, the device comprising: a receiving unit for obtaining a voice sample from the subject; an extraction unit for extracting at least one voice biomarker feature value from the voice sample for at least one predetermined voice biomarker feature; a determining unit for determining the blood glucose level for the subject based on the at least one voice biomarker feature value and a blood glucose level prediction model; and an output unit for outputting the blood glucose level or an output based on the blood glucose level for the subject.