BioStampRC Customer Login
OUR BLOG

MC10 Pulse

How Multi-Modal Sensors Outperform Wrist-Worn Wearables

Friday, April 07, 2017   • Posted by Erika Vázquez

In an earlier blog, I discussed the value of not limiting data collection sources to the wrist. Though several devices offer wrist-captured physiological signals, this data is limited in scope and not representative of the body’s full spectrum of movement. To highlight this point, we placed BioStampRC Sensors in three locations on a volunteer’s body — the wrist, chest and thigh. Each Sensor captured a distinct accelerometer signal. Once the Sensors were applied, the subject completed a series of activities.

Read on to learn what our experiment demonstrated about the value of multi-location sensing.

Study Summary

Activity 1: Walking — Subject walked in a straight line for 22 seconds.

Data from Sensor Placed on Subject’s Wrist:

Wrist Sensor Data Walking

Data from Sensor Placed on Subject’s Chest:

Chest Sensor Data Walking

Data from Sensor Placed on Subject’s Thigh:

Thigh Sensor Data Walking

Activity 2: Sit to Stand — Subject was instructed to sit in a straight back chair, stand up, and walk for 3 meters.

Data from Sensor Placed on Subject’s Wrist:

Wrist Sensor Data Sit to Stand

Data from Sensor Placed on Subject’s Chest:

Chest Sensor Data Sit to Stand

Data from Sensor Placed on Subject’s Thigh:

Thigh Sensor Data Sit to Stand

Activity 3: Stationary Bike — Subject was instructed to ride on a stationary recumbent bike at a resistance of 1, take a break, and continue at a resistance of 10.

Data from Sensor Placed on Subject’s Wrist:

Wrist Sensor Data Stationary Bike

Data from Sensor Placed on Subject’s Chest:

Chest Sensor Data Stationary Bike

Data from Sensor Placed on Subject’s Thigh:

Thigh Sensor Data Stationary Bike

The accelerometry motion signature from each of these activities varies widely based on the body placement of the Sensors. The precision of devices offering location agnostic sensing allows for not only targeted, accurate data collection from a specific area but also a more thorough understanding of the body’s movement when multiple body locations are recorded simultaneously.

The Benefits of Multi-Modal Sensors

Just as multiple location sources for data collection lead to greater data accuracy, examining more than one physiological signal leads to a more conclusive assessment of the body’s motion.

For example, when comparing the movement of someone walking through water to walking on land, the linear motion data from the subject’s leg would be nearly indistinguishable. However, by adding an electromyography (EMG) sensor to the equation, the level of exertion necessary to complete both activities would vary widely. The water walking would require a significant amount of muscle activation when compared to the regular walking.

In a similar experiment, an MC10 subject wore a BioStampRC Sensor on his thigh, recording both accelerometer and EMG signals as he pedaled on a stationary bike. The subject pedaled at a resistance of 1, paused to take a break, and then continued pedaling at a resistance of 10.

Data from Sensor Placed on Subject’s Thigh:

Multi-Modal Sensor Data

When looking at the data from the accelerometer, the movement during the resistance 1 and resistance 10 pedaling is indistinguishable. But after adding the EMG data, there is a clear increase in muscle activity when the subject pedals at a higher resistance.

By using multi-modal sensors, researchers can more confidently assess a subject’s movement and experience by identifying differences in motions that would otherwise go unseen.

Get Involved