Real-World M3-BREATHE: Toward Multimodal Mobile Monitoring of Behaviour, Respiration, and Exposures for Treatment and Health Evaluation
June 5th, 2026
M3-BREATHE is a mobile health (mHealth) system that integrates wearable sensors to monitor respiration, behaviour, and environmental exposures for personalized treatment and health evaluation in the real world.
M3-BREATHE application to integrate behavior, respiration, and exposure, examine how mobility influences real-time health risks and outcomes. Existing studies use standing locations and missing mobility. This protocol captures high-resolution synchronized individual-level data across behavior, physiology, and exposures.
To begin, wear the sensor integrated shirt or chest strap directly against the skin beneath the clothing to ensure optimal contact. Adjust the fit so that it is snug and comfortable without constraining breathing or movement. Confirm stable skin contact by checking that the sensor does not slip during deep inhalation.
Ensure the sensor is fully charged and pair it with the smartphone via Bluetooth for real-time monitoring. Confirm successful binding and automatic synchronization of multimodal data streams between the smartphone and the linked devices through the dedicated application. After choosing the heart rate sensor, verify that the heart rate and RR interval or RRI values are being dynamically recorded on the smartphone.
To set up the smart wristband, choose the non-dominant wrist and secure the device proximal to the ulnar styloid process. Adjust the strap to ensure proper sensor skin contact while allowing insertion of one fingertip. Confirm that the photoplethysmography or PPG sensor maintains constant and flush contact with the skin.
After opening the companion application, connect the wristband via Bluetooth to the smartphone. Verify the pairing status and complete the band setup within the application. Set the location access to always and toggle on to enable the motion and fitness option.
Monitor the cardiovascular metrics, including heart rate or HR in beats per minute, heart rate variability or HRV in milliseconds, blood pressure or BP in millimeters of mercury, and electrocardiogram or ECG in millivolts. Similarly, monitor the respiratory rate along with other parameters like heart rate and RRI. Under the health measurements menu, select SpO2 to visualize the variations in blood oxygen saturation percentage throughout the day or over weeks, months, and years.
In the stats menu, choose active calories to see the movement breakdown and the real-time movement data across daily, weekly, monthly, or yearly trends. Track the daily, weekly, or monthly peripheral skin temperatures in degrees Celsius by selecting the body temperature option under the health measurements menu and motion data from the accelerometer. Confirm that all metrics, including motion data, sleep, weight management, cardiovascular, respiratory, and temperature data, along with a summary, are visible.
For the environmental sensor setup, connect environmental sensing devices to the portable power bank using appropriate USB cables for 24-hour operations. Place the multi-air pollutant monitor, noise, and temperature sensors in the fishnet side pocket of the participant's backpack. Monitor air pollution parameters like particulate matters.
PM 2.5 values in micrograms per cubic meters. Assess the noise pollution by measuring the live volume levels in decibels and temperature from the temperature sensor in degrees Celsius. Next, attach the light exposure tracker to the shoulder strap facing outward to measure the illuminance in luxe and circadian light A or CLA values as zero to one.
Instruct the participant to carry the backpack during daily activities for 24 hours and charge the power bank every night. Confirm that all sensors are powered on and transmitting real-time environmental data, including air particulate matter and formaldehyde concentration. Enable Bluetooth and set location access to always in the menu.
Configure the application to record GPS location data at one-minute intervals and sync sensor data every five minutes. In the device's interface, select End Exercise to complete data synchronization. In the device's interface, select the activity records to confirm that all daily tracking reminders are set to on.
Select heart rate option to verify that continuous heart rate monitoring is enabled. Under emotional wellbeing, confirm that emotion stress records is active. Then enable both notifications and cloud data sync.
Under blood oxygen levels, verify that the automatic SpO2 is enabled. Review a short test recording to ensure data completeness and accuracy, checking for correct timestamping and plausible values across all streams. Record the different modes of daily movements and travel episodes under the action center, along with GPS trajectories.
Provide instructional materials to the participants, including manuals and videos for device usage. Assess the demographic data parameters like gender, age, weight, height, and marital status of the participants. Instruct the participant to wear physiological sensors and carry environmental devices continuously for 24 hours.
Remind the participant to recharge the power bank daily to maintain uninterrupted operation and data. Monitor the data stream through manual checks or via automated backend alerts. Under the calorie records, enter the diet log for different meals, including breakfast, lunch, dinner, or an extra meal.
Average PM 2.5 concentrations and subjective wellbeing scores varied significantly across daily activities with exposure differing markedly between indoor versus outdoor and sedentary versus active settings. The highest average exposure exceeding 20 micrograms per cubic meter was observed during sleep, surfing the internet, entertainment, studying, and taking a walk. Mid-range exposure concentrations in the range of 15 to 20 micrograms per cubic meter were found for activities such as work, physical activity, caregiving, and trips.
The lowest exposures in the range of 5 to 15 micrograms per cubic meter were recorded during personal affairs, seeing a doctor, dining, and personal care. The cognitive subjective wellbeing scores showed less variability across activities, ranging narrowly between 1.7 and 2.3 on a one to five dissatisfaction scale. The lowest satisfaction scores were associated with seeing a doctor, trips, personal affairs, and work.
The highest satisfaction scores were linked to physical activity, entertainment, and social activity. The activity types associated with higher PM 2.5 exposure concentrations did not always correspond to lower cognitive subjective wellbeing, suggesting decoupling patterns between PM 2.5 exposure and self-reported cognitive subjective wellbeing. M3-BREATHE enables real-time measurements of personalized exposure to higher doses of environmental pollutants and the physiological responses during daily activities and travel.
The key challenge is reliable device collaboration, synchronization, and sustained participant compliance in the real world long-term monitoring. Future work can integrate AI interventions, genomics, and core city deployments to study course pathways and improve precision around the house.
Overview
The M3-BREATHE protocol introduces a multimodal mobile monitoring system that integrates mHealth devices, portable environmental sensors, and GPS-enabled smartphone applications to capture synchronized data on individual behavior, respiratory function, physiological parameters, and environmental exposures. The system processes data through a mobile application to calculate physiological parameters and incorporate urban environmental data, enabling a machine learning-based alert system for actionable insights into respiratory health, daily activities, mobility patterns, and exposure to risk factors. The protocol provides comprehensive guidance on hardware setup, software configuration, data acquisition, and analytical workflows for deployment in both controlled clinical experiments and real-world contexts. M3-BREATHE establishes a scalable, non-invasive, and cost-efficient approach to longitudinal environment-mobility-health monitoring, linking individual-level precision health with population-scale health intelligence.
Key Study Components
Area of Science
- Mobile health (mHealth) technologies
- Environmental exposure monitoring
- Physiological and behavioral signal monitoring
- Machine learning for health risk assessment
Background
- Continuous real-time monitoring of physiological and behavioral signals holds transformative potential for personalized health risk assessment.
- Integration with detailed environmental exposure measurement enables more effective treatment and intervention strategies.
- Existing approaches often lack synchronization of multimodal data streams for comprehensive environment-mobility-health analysis.
- There is a need for scalable, non-invasive, and cost-efficient platforms for longitudinal health monitoring in real-world settings.
Purpose of Study
- To introduce a multimodal mobile monitoring system (M3-BREATHE) for capturing and integrating data on behavior, respiration, and environmental exposures.
- To process synchronized data streams using physiological parameter calculation and urban environmental data integration.
- To deploy a machine learning-based alert system for actionable insights into respiratory health, daily activities, mobility patterns, and risk factor exposure.
- To provide comprehensive guidance for deploying M3-BREATHE in controlled clinical experiments and real-world contexts.
Methods Used
- Integration of mobile health devices and portable environmental sensors with GPS-enabled smartphone applications.
- Synchronized, high-resolution data acquisition of daily activities, travel, respiratory function, physiological parameters, and environmental exposures (air pollution, noise, light, heat, greenness).
- Processing of data through a smart mobile phone application incorporating physiological parameter calculation (e.g., physical activity intensity, respiratory rate) and external urban environmental data (e.g., urban green spaces, street view maps).
- Deployment of a machine learning-based alert system to generate actionable insights for respiratory health, daily activities, mobility patterns, and exposure to risk factors.
Main Results
- The M3-BREATHE system successfully generates synchronized, high-resolution data streams from multimodal sources.
- Physiological parameter calculation and urban environmental data integration enable context-aware health monitoring.
- The machine learning-based alert system provides actionable insights into respiratory health, daily activities, mobility patterns, and environmental risk factor exposure.
- The protocol supports deployment in both controlled clinical experiments and real-world contexts, demonstrating scalability and practical utility.
Conclusions
- M3-BREATHE establishes a scalable, non-invasive, and cost-efficient approach to longitudinal environment-mobility-health monitoring.
- The platform links individual-level precision health in everyday life with long-term health intelligence at the population scale.
- By integrating behavior, respiration, and environmental exposure data, M3-BREATHE advances personalized health risk assessment and intervention effectiveness.
- The comprehensive guidance provided enables reproducible deployment of the system across diverse research and clinical settings.
Frequently Asked Questions
Executive Industry Relevance
The M3-BREATHE platform enables continuous, real-time monitoring of behavior, respiration, and environmental exposures, supporting longitudinal health monitoring in real-world settings. By integrating multimodal data streams through a machine learning-based alert system, it provides actionable insights for respiratory health and mobility patterns. This scalable, non-invasive approach links individual-level precision health with population-scale health intelligence, advancing personalized risk assessment and intervention strategies.
Strategic Applications in Biopharma R&D
Early Discovery & Target Validation
- Scientific Value: Enables interrogation of behavioral and respiratory phenotypes in naturalistic environments to support target hypothesis testing.
- Operational Value: Provides synchronized, high-resolution data streams for de-risking mechanistic assumptions about respiratory function and environmental interactions.
Screening & Assay Development
- Scientific Value: Generates quantitative physiological and environmental readouts suitable for assay standardization and reproducibility testing.
- Operational Value: Supports preparation of validated biological systems for downstream compound screening by capturing context-dependent variability.
Translational & Preclinical Research
- Scientific Value: Facilitates disease-relevant system modeling by linking individual-level exposure and response data to population health trends.
- Operational Value: Enables risk-adjusted advancement decisions through longitudinal environment-mobility-health monitoring in translational workflows.
Pipeline & Workflow Integration
The M3-BREATHE system integrates into discovery biology by enabling hypothesis testing through real-world behavioral and respiratory phenotyping, supporting early target validation.
- Discovery Biology: Supports hypothesis testing and pathway clarification by capturing synchronized data on behavior, respiration, and environmental exposures in everyday life.
- Screening: Enhances assay readiness by providing reproducible, quantitative physiological and environmental outputs for compound evaluation.
- Analytics: Delivers machine learning-derived insights and physiological parameter calculations that enable cross-condition comparison and predictive modeling.
- Translational Research: Connects discovery to preclinical continuity by aligning individual-level precision monitoring with population-scale health intelligence.
- Enterprise Reuse: Functions as a reusable platform for longitudinal monitoring across therapeutic areas, reducing reliance on single-use techniques.
Operational & Enterprise Impact
- Scientific Value: Increases predictive confidence by reducing mechanistic ambiguity in respiratory health and environmental risk factor interactions.
- Operational Value: Ensures standardization, reproducibility, and scalability through synchronized multimodal data acquisition and processing workflows.
- Strategic Value: Improves go/no-go decisions by enabling earlier detection of biological risk and enhancing capital efficiency in discovery programs.
- Portfolio Impact: Supports risk-adjusted prioritization through longitudinal monitoring that informs advancement decisions based on real-world health trajectories.
Implementation Considerations
- Requires expertise in mobile health technologies, environmental sensing, and data integration for proper system deployment.
- Depends on instrumentation including mHealth devices, portable environmental sensors, and GPS-enabled smartphones for synchronized data capture.
- Necessitates cross-team standardization of data acquisition protocols and analytical workflows to ensure reproducibility across sites.
- Involves adaptation considerations when applying the system across diverse model systems or environmental contexts.
- Includes practical limitations such as device battery life, data synchronization challenges, and environmental sensor calibration needs as noted in the protocol.
