Design and Evaluation of Smart Glasses for Food Intake and Physical Activity Classification
February 14th, 2018
11.2K views
This study presents a protocol of designing and manufacturing a glasses-type wearable device that detects the patterns of food intake and other featured physical activities using load cells inserted in both hinges of the glasses.
Overview
This study presents a protocol for designing and manufacturing a glasses-type wearable device that detects patterns of temporalis muscle activities during food intake and other physical activities. This method aims to provide an objective way to monitor ingestible behaviors.
Key Study Components
Area of Science
- Neuroscience
- Wearable technology
- Behavioral monitoring
Background
- Monitoring food intake is crucial for understanding eating behaviors.
- Existing methods may lack objectivity and automation.
- Wearable devices can enhance data collection in real-time.
- The temporalis muscle's activity is relevant for food intake detection.
Purpose of Study
- To design a wearable device that detects muscle activity during eating.
- To automate the monitoring of food intake patterns.
- To provide an objective measurement tool for researchers.
Methods Used
- Design and manufacturing of glasses-type device.
- Integration of load cells in the device hinges.
- Data collection procedures conducted by trained personnel.
- Demonstration of the manufacturing process by a mechanical engineer.
Main Results
- The device successfully detects temporalis muscle activity.
- Patterns of food intake can be monitored automatically.
- Objective data collection enhances research capabilities.
- Demonstrated feasibility of the manufacturing process.
Conclusions
- The glasses-type device is a promising tool for behavioral monitoring.
- It provides an innovative approach to studying food intake.
- Future research can build on this technology for broader applications.
Frequently Asked Questions
What is the main advantage of the wearable device?
The device allows for automatic and objective detection of muscle activity during food intake.
Who conducted the data collection?
Data collection was performed by Hak Yung Gu, a team member at the company.
What is the role of the mechanical engineer in this study?
Wonjoon Oh, the mechanical engineer, demonstrated the manufacturing procedure of the device.
How does this device contribute to neuroscience research?
It provides a new method for monitoring ingestible behaviors, enhancing data accuracy.
What type of data does the device collect?
The device collects data on temporalis muscle activity related to food intake.
Can this technology be applied to other physical activities?
Yes, the device is designed to monitor various physical activities beyond food intake.
Executive Industry Relevance
Objective, real-time monitoring of ingestive behaviors is a critical challenge in metabolic and behavioral research pipelines. The integration of wearable sensor technology, such as smart glasses detecting temporalis muscle activity, enables automated, quantitative assessment of food intake and related physical activities. This capability supports translational research and portfolio decisions by providing high-fidelity behavioral endpoints for early discovery and preclinical studies.
Strategic Applications in Biopharma R&D
Early Discovery & Target Validation
- Enables objective measurement of ingestive behaviors for hypothesis testing in metabolic and CNS research.
- Supports functional validation of behavioral endpoints linked to target engagement.
- Facilitates mechanistic de-risking by distinguishing food intake from confounding activities.
Screening & Assay Development
- Provides standardized, reproducible data streams for behavioral phenotyping assays.
- Delivers quantitative outputs suitable for algorithmic classification and downstream analytics.
- Enables scalable, high-throughput screening of interventions affecting ingestive behavior.
Translational & Preclinical Research
- Aligns behavioral monitoring with disease-relevant endpoints in metabolic and neurological models.
- Supports continuity from discovery through preclinical validation by enabling objective, automated data capture.
- Improves predictive confidence in translational biomarker development for ingestive disorders.
Pipeline & Workflow Integration
This wearable sensing platform fits within the early discovery to preclinical continuum, enabling robust behavioral phenotyping and target validation workflows.
- Discovery Biology: Supports hypothesis-driven interrogation of ingestive mechanisms and behavioral pathways.
- Screening: Provides reproducible, quantitative behavioral readouts for compound or intervention evaluation.
- Analytics: Generates feature vectors and classification outputs for statistical comparison across experimental conditions.
- Translational Research: Bridges preclinical and clinical endpoints by enabling objective, real-world behavioral monitoring.
- Enterprise Reuse: Offers a modular, adaptable platform for diverse behavioral and physiological monitoring applications.
Operational & Enterprise Impact
- Scientific Value: Increases predictive confidence and reduces ambiguity in behavioral endpoint measurement.
- Operational Value: Standardizes data acquisition and enhances reproducibility across studies and teams.
- Strategic Value: Informs go/no-go decisions with objective, high-resolution behavioral data.
- Portfolio Impact: Enables risk-adjusted prioritization of assets targeting ingestive and metabolic pathways.
Implementation Considerations
- Requires expertise in wearable device fabrication and sensor integration.
- Demands robust data acquisition and wireless transmission infrastructure.
- Necessitates cross-team standardization of behavioral protocols and data analysis pipelines.
- Adaptable to various model systems with appropriate frame sizing and fit adjustments.
- Practical limitations include the need for subject compliance and potential confounding from non-ingestive facial movements.
Biopharma FAQs
Why does null hypothesis testing matter for food intake classification?
Null hypothesis testing ensures that observed differences in temporalis muscle activity patterns are statistically significant, supporting robust target validation and reducing false positives in ingestive behavior studies.
How does independent variable isolation fit in activity detection protocols?
Isolating variables such as specific physical activities (chewing, walking, talking) allows for precise attribution of sensor signal changes, enhancing the reliability of behavioral classification in discovery workflows.
What do quantitative dependent variable measurements enable in this system?
Quantitative force signal measurements from the load cell modules enable objective, algorithm-driven classification of ingestive and non-ingestive activities, supporting reproducible and scalable data analysis.
Why are replication requirements critical for cross-functional behavioral studies?
Replication across multiple sessions and subjects ensures that behavioral classification outputs are robust and generalizable, facilitating collaboration between discovery, analytics, and translational teams.
What statistical analysis capabilities are required before implementation?
Robust statistical analysis, including feature extraction and classification accuracy metrics such as F1 score, is essential to validate the system's performance and inform go/no-go decisions in R&D pipelines.
Chapters in this video
0:04
Title
0:59
The Sensor-integrated Circuit Modules
1:38
3D Printing of the Glasses Frame
2:39
Assembly of the Glasses
3:47
Development of the Data Acquisition System
4:28
Data Collection from a User Study
6:44
Results: Representative Temporal Data Collected for Different Activities
7:25
Conclusion
