element14 Community
element14 Community
    Register Log In
  • Site
  • Search
  • Log In Register
  • Members
    Members
    • Benefits of Membership
    • Achievement Levels
    • Members Area
    • Personal Blogs
    • Feedback and Support
    • What's New on element14
  • Learn
    Learn
    • Learning Center
    • eBooks
    • STEM Academy
    • Webinars, Training and Events
    • Learning Groups
  • Technologies
    Technologies
    • 3D Printing
    • FPGA
    • Industrial Automation
    • Internet of Things
    • Power & Energy
    • Sensors
    • Technology Groups
  • Challenges & Projects
    Challenges & Projects
    • Design Challenges
    • element14 presents
    • Project14
    • Arduino Projects
    • Raspberry Pi Projects
    • Project Groups
  • Products
    Products
    • Arduino
    • Dev Tools
    • Manufacturers
    • Raspberry Pi
    • RoadTests & Reviews
    • Avnet Boards Community
    • Product Groups
  • Store
    Store
    • Visit Your Store
    • Choose Another Store
      • Europe
      •  Austria (German)
      •  Belgium (Dutch, French)
      •  Bulgaria (Bulgarian)
      •  Czech Republic (Czech)
      •  Denmark (Danish)
      •  Estonia (Estonian)
      •  Finland (Finnish)
      •  France (French)
      •  Germany (German)
      •  Hungary (Hungarian)
      •  Ireland
      •  Israel
      •  Italy (Italian)
      •  Latvia (Latvian)
      •  
      •  Lithuania (Lithuanian)
      •  Netherlands (Dutch)
      •  Norway (Norwegian)
      •  Poland (Polish)
      •  Portugal (Portuguese)
      •  Romania (Romanian)
      •  Russia (Russian)
      •  Slovakia (Slovak)
      •  Slovenia (Slovenian)
      •  Spain (Spanish)
      •  Sweden (Swedish)
      •  Switzerland(German, French)
      •  Turkey (Turkish)
      •  United Kingdom
      • Asia Pacific
      •  Australia
      •  China
      •  Hong Kong
      •  India
      •  Korea (Korean)
      •  Malaysia
      •  New Zealand
      •  Philippines
      •  Singapore
      •  Taiwan
      •  Thailand (Thai)
      • Americas
      •  Brazil (Portuguese)
      •  Canada
      •  Mexico (Spanish)
      •  United States
      Can't find the country/region you're looking for? Visit our export site or find a local distributor.
  • Translate
  • Profile
Startupbootcamp IoT
  • Technologies
  • Business of Engineering
  • Startupbootcamp IoT
  • More
  • Cancel
Startupbootcamp IoT
Blog Mechion | Using Pytorch Deep Learning to Detect Early Signs of Dementia
  • Blog
  • Forum
  • Documents
  • Events
  • Polls
  • Files
  • Members
  • Mentions
  • Sub-Groups
  • Tags
  • More
  • Cancel
  • New
Startupbootcamp IoT requires membership for participation - click to join
Blog Post Actions
  • Subscribe by email
  • More
  • Cancel
  • Share
  • Subscribe by email
  • More
  • Cancel
Group Actions
  • Group RSS
  • More
  • Cancel
Engagement
  • Author Author: jlucas
  • Date Created: 7 Nov 2018 3:47 PM Date Created
  • Views 421 views
  • Likes 2 likes
  • Comments 0 comments
  • business of engineering
  • raspberry pi zero w
  • startup bootcamp
  • raspberry pi zero
  • artificial interlligence
  • pi-zero
  • raspberri pi
  • pi zero w
  • startup stories
  • mechion
  • deep learning
  • nvidia
  • pytorch
  • startupbootcamp iot
Related
Recommended

Mechion | Using Pytorch Deep Learning to Detect Early Signs of Dementia

jlucas
jlucas
7 Nov 2018

For the third year in a row, Farnell element14 is proud to be an official partner of Startupbootcamp IoT, an intensive three-month acceleration program for tech-based startups. We recently traveled to Startupbootcamp IoT's London headquarters to meet the participating startups for the 2018 cohort.

 

You don't have permission to edit metadata of this video.
Edit media
x
image
Upload Preview
image

 

Introducing Mechion

 

Mechion is a UK-based IoT startup that uses computer vision and deep learning to detect the visual indicators and emotional behaviour patterns that are often an early sign of dementia.

 

How it works

 

For many elderly and vulnerable people, signs of the onset and progress of dementia can only be tracked through hospitals and specialised clinics. Mechion uses an unobtrusive hardware device that can be installed in the home to detect and record key indicators of concern. This is then collected as non-visual meta data and sent to a back-end platform where the data is processed into reports, analytics and insights which can be shared with the user's family and medical professionals. The company is currently preparing to begin clinical trials with several hospitals to explore how this information can be used to predict dementia and to track whether a patient's condition is getting worse.

 

The Hardware

 

Initial versions of Mechion's hardware device used an NVIDIA Tegra TX2 module with custom-built carrier boards and Intel D414 depth-sensing cameras to collect data such as facial expressions, movement and gait analysis. However, due to the high cost of the platform, they are now working with the Raspberry Pi 0W and using convolutional neural networks to provide image classification and segmentation. Using Raspberry Pi as an end-point, data is collected and sent to a cloud platform for processing.

 

The Software

 

Mechion's software incorporates custom-built neural architectures and state-of-the-art existing models such as Google's ResNet, which can be trained to process the data collected by the hardware device. This is achieved using the PyTorch deep learning platform and NVIDIA's CUDA compute capabilities.

 

The USP

 

According to Mechion co-founder Dr. Alexander Gkiokas, the key selling point of his product is that it's a very passive system that in no way interferes with the user's life. Once installed, it collects data round the clock, learning from the user's unique behaviour to generate highly specific insights and analytics in a way that is absolutely safe and private. No images or video footage is collected, the reports purely look at metadata based on information collected by the hardware. This means that the user can potentially enjoy more independence and autonomy during the early stages of dementia while also allowing for family and healthcare professionals to respond more quickly should more serious issues arise.

 

Looking to the future

 

In the short term, the Mechion team are keeping the focus on dementia, working to improve the accuracy with which they can pinpoint the signs and detect potential issues. One of the biggest challenges they currently face is the logistics of processing the huge amounts of data required to generate their reports. Their current prototype produces around 20GB of data per day, per person. They are currently working on an algorithm to reduce this amount of data so that it can be processed as close to real time as possible.

 

Once these issues have been resolved, they also have ambitions to look at adapting their technology to help identify early signs of infection, which can often develop into major issues and create secondary problems in older people.

 

Joining the Startupbootcamp IoT Program...

 

"We were a little sceptical at first" admits Dr. Gkiokas about his reasons for joining the Startupbootcamp IoT program. "However, our attitude has chained dramatically since we actually started. From the mentoring and workshops to the constant feedback, advice and constructive criticism, it's really driven us to refine our product and our approach not only to the business side of running a startup, but also the technical strategy. It's a truly unique experience, and I think it's going to help us to become more feasible and sustainable as a business."

 

We'll be continuing to check in with Mechion and the other Startupbootcamp IoT 2018 participants over the coming weeks, so keep checking this space for more updates. You can also learn more about Mechion on their official website and on the Startupbootcamp website.

  • Sign in to reply
element14 Community

element14 is the first online community specifically for engineers. Connect with your peers and get expert answers to your questions.

  • Members
  • Learn
  • Technologies
  • Challenges & Projects
  • Products
  • Store
  • About Us
  • Feedback & Support
  • FAQs
  • Terms of Use
  • Privacy Policy
  • Legal and Copyright Notices
  • Sitemap
  • Cookies

An Avnet Company © 2023 Premier Farnell Limited. All Rights Reserved.

Premier Farnell Ltd, registered in England and Wales (no 00876412), registered office: Farnell House, Forge Lane, Leeds LS12 2NE.

ICP 备案号 10220084.

Follow element14

  • Facebook
  • Twitter
  • linkedin
  • YouTube