Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Evaluate Person's Pulse Data with Machine Learning

Gabriela de Queiroz
August 21, 2018
69

Evaluate Person's Pulse Data with Machine Learning

Gabriela de Queiroz

August 21, 2018
Tweet

Transcript

  1. Evaluate Person's Pulse Data
    with Machine Learning
    Gabriela de Queiroz
    Senior Developer Advocate, IBM
    [email protected]

    View full-size slide

  2. CALL FOR CODE
    2
    • Worldwide, multi-year initiative that
    inspires developers to solve pressing global
    problems with sustainable software
    solutions.
    • Create applications with IBM Cloud
    technology that improve disaster
    preparedness and relief, build resilient
    communities, and safeguard the health and
    well-being of individuals and institutions.
    https://developer.ibm.com/callforcode/

    View full-size slide

  3. Agenda
    3
    In this 30-minute online meetup you will
    learn:
    - How to build a mobile web app to check a
    person's pulse rate
    - How to evaluate a person's pulse data with
    Machine Learning
    - How to leverage pulse data
    during emergencies and natural disasters

    View full-size slide

  4. 4
    https://developer.ibm.com/code/patterns/develop-web-based-mobile-health-app-uses-machine-learning/

    View full-size slide

  5. DISCLAIMER: This application is used for
    demonstrative and illustrative purposes
    only and does not constitute an offering that
    has gone through regulatory review. It is not
    intended to serve as a medical application.
    There is no representation as to the
    accuracy of the output of this application
    and it is presented without warranty.
    5

    View full-size slide

  6. General Workflow
    6

    View full-size slide

  7. Clone/Download the github repo:
    https://github.com/IBM/pulse-iot-wml-mobile-health
    9

    View full-size slide

  8. Steps:
    1.Create the Node.js app.
    2.Set up the Watson Machine Learning service in IBM
    Cloud.
    3.Set up Watson IoT Platform in IBM Cloud.
    4.Set up the machine learning model in IBM Watson Studio.
    5.Try it out on your smartphone.
    10

    View full-size slide

  9. Step 1: Create a
    Node.js
    (Node.js Cloudant DB Web Starter)
    11

    View full-size slide

  10. 1.1: Create a Node.js and Cloudant
    DB
    12

    View full-size slide

  11. This will will create the app and the CloudantNoSQLDB
    App
    CloudantNoSQLDB

    View full-size slide

  12. Step 2: Set up the Watson
    Machine Learning service in IBM
    Cloud.
    15

    View full-size slide

  13. 2.1: Create a Machine Learning service
    16

    View full-size slide

  14. • Create a project
    • Add new Watson ML model

    View full-size slide

  15. 2.2: Save the credentials
    18

    View full-size slide

  16. Step 3: Set up Watson IoT
    Platform in IBM Cloud.
    19

    View full-size slide

  17. • Create a project
    • Add new Watson ML model

    View full-size slide

  18. • Add it by binding from within your node.js app (click on "+ Add A Connection" to add this service to your app)
    Make sure the service is binded

    View full-size slide

  19. Step 4: Set up the Machine Learning
    Model in IBM Watson Studio.
    22

    View full-size slide

  20. 4.1: Create a new project
    23

    View full-size slide

  21. 4.2: Create a ML model
    24
    Under Assets, click on “New Watson Machine Learning Model”

    View full-size slide

  22. Define the model details:
    1) Give it a name
    2) Associate a Spark Service
    3) Associate a Machine Learning Service

    View full-size slide

  23. 4.3: Associate a Spark instance
    26

    View full-size slide

  24. 4.4: Associate a Machine Learning service instance
    27

    View full-size slide

  25. 4.5: Create a notebook
    29
    Under Assets > Notebooks > New notebook

    View full-size slide

  26. 4.6: Import Notebook and Define the language and Spark version
    30
    You can import the ipython notebook "Pulse Rates.ipynb" (in the repo)

    View full-size slide

  27. 4.7: Run the ipython notebook "Pulse Rates.ipynb"
    31
    Don’t forget to fill your Watson ML Credentials!!

    View full-size slide

  28. Step 5: Deploy the App
    32

    View full-size slide

  29. 5.1: After the model is deployed get the credentials and
    save to .env
    33

    View full-size slide

  30. 5.2: Run each line by line the content of the deploy.sh
    34

    View full-size slide

  31. The app is ready!
    35

    View full-size slide

  32. DOC ID / Month XX, 2018 / © 2018 IBM Corporation
    How could I leverage the
    pulse rate app during
    emergencies and natural
    disasters?
    36

    View full-size slide

  33. DOC ID / Month XX, 2018 / © 2018 IBM Corporation 37
    Source: https://news.stjohnvic.com.au/first-aid-hypothermia-treatment/

    View full-size slide

  34. Cave Rescue
    38

    View full-size slide

  35. Landslides
    39

    View full-size slide