Edge Impulse logo
Blog Post

Utilizing On-Device Machine Learning for Health and Safety Monitoring

Artificial Intelligence, Machine Learning, Embedded DevicesIn this article, we will explore an emerging area where wearable and machine learning technology are driving innovation: industrial safety.

Jed Huang, Sheena Patel

July 13, 2022

The wearable technology market is rapidly growing, driven by consumer and industry trends like personalized health, industrial safety, and worker monitoring. It is estimated to grow at a CAGR of around 18.5% from 2022 to 2028. In 2021, the wearable technology market was valued at around $115.8B USD and it is projected to reach $380.5B by 2028.

This technology is no longer just smart bands and watches for fitness tracking, wearable AI and machine learning are now moving into new form factors like rings, straps, and clothing and enable deeper insights into health and wellness, allowing users to improve aspects of their everyday lives like sleep, activity, stress, and fatigue. 

In this article, we will explore an emerging area where wearable and machine learning technology are driving innovation: industrial safety.

Read the Industrial Safety: SlateSafety sectionIndustrial Safety: SlateSafety

SlateSafety is a technology company that began its journey providing real-time  physiological monitoring to first responders and industrial workers to prevent injuries in the field.

The company now also provides solutions for the industrial safety markets, like worker safety monitoring for large industrial conglomerates. In June of 2022, the AG of New Jersey led a coalition for a new set of rules that would require employers to submit more reports to the Occupational Safety and Health Administration (OSHA) around worker injuries. [1] Given the increased regulation, companies with employees working physically and mentally strenuous jobs must now take extra precaution to monitor workers well-being and implement preventative measures as workers become exerted. 

SlateSafety helps large industrial companies keep their workers safe with a wearable band called the BioTrac Band that monitors workers’ vitals like their heart rate, exertion levels, body temperature, and more, with a central dashboard that can be used by a team leader or supervisor. 

Edge Impulse worked directly with the SlateSafety team to develop a heat exhaustion predictive algorithm using biometric data being collected off of the BioTrac Band. This algorithm detects risk of heat exhaustion five to 10 minutes in advance and can send a notification to the wearer and their supervisor for preventative action.

Read the A Scalable Approach to Health Machine Learning Engineering sectionA Scalable Approach to Health Machine Learning Engineering

Edge Impulse works with companies in the industrial safety and health devices space like Oura, SlateSafety, and Knowlabs on their full data engineering and machine learning algorithm development process. A platform approach for this process ensures that data is labeled properly, and that algorithms are built against data being collected from gold standard tests to ensure accuracy. Furthermore, algorithms can be tested and improved over time, and even personalized. This is especially critical for health-related algorithms to detect conditions such as fatigue or stress levels where variations between physiology may exist.

 Screenshot from the Edge Impulse platform showing feature importance during the health algorithm development process.

Through our enterprise platform, APIs, and solutions offering, Edge Impulse helps wearables and health device companies with the following:

  • Bringing new algorithms to market faster
  • Building robust, accurate algorithms on par with gold standard medical-grade tests
  • Improving, personalizing, and updating algorithms over time
  • Device data streaming integration and automation of visualizations with new data input

Read the Deploying Health Algorithms  sectionDeploying Health Algorithms 

The Edge Impulse platform can help deploy algorithms anywhere — whether that is to the wearable hardware, a cell phone or even to the cloud. This flexibility allows health device companies to make tradeoffs on how they want to architect the data flow — from data collection, feature extraction, and algorithm classification to algorithm deployment. 

As new AI accelerated hardware releases into the market, from companies like Silicon Labs, Alif Semiconductor and Syntiant, wearables and devices manufacturers hardware teams may choose to design in these processors. Edge Impulse has the most robust ecosystem of hardware partners — both general purpose and neural accelerated processors that range from low-power MCUs to more robust processors with higher compute available.

 Edge Impulse algorithms can deploy anywhere, whether to supported hardware, custom hardware or boards, cell phones / gateways or the cloud.

In summary, the Edge Impulse team has the experience and tooling to help wearables and device companies bring new and innovative health and performance algorithms, and products to market faster.

Please feel free to reach out or book an exploratory call to learn more about how we can help your team build robust, gold standard health algorithms here.

If you’re interested in learning more, check out the links below:

  1. New OSHA Rules in New Jersey for Decreasing Worker Injuries
  2. SlateSafety Heat Exhaustion Predictive Algorithm Video
  3. SlateSafety Heat Exhaustion Predictive Algorithm Case Study

Subscribe

Are you interested in bringing machine learning intelligence to your devices? We're happy to help.

Subscribe to our newsletter