Salvando vidas com dispositivos semi-conectados

Apresentada na trilha Saúde 4.0 do TDC Porto Alegre 2019

Transcript

1. Salvando vidas com
   dispositivos semi-conectados
   Julio Faerman
   @faermanj
   julio.faerman@nubego.io
   

   View Slide

2. Julio Faerman
   @faermanj
   ● Independent Software Engineer
   specialized in Cloud Computing
   ● Can code but not solder
   ● Before: Amazon Web Services
   Developer Relations & Training
   ● Previously: Red Hat, Borland and
   Enterprise Computing in Brazil
   ● Today: Own thoughts and
   references of a maker
   

   View Slide

3. “We are
   no longer
   patients
   we are
   consumers
   of health.”
   AWS re:Invent 2014 | Philips Uses AWS to Analyze 15 PB of Patient Data
   

   View Slide

4. View Slide

5. View Slide

6. View Slide

7. c
   Brazil’s healthtech sector
   is new hot spot
   Manoel Lemos
   c
   

   View Slide

8. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
   S U M M I T
   

   View Slide

9. Engineering for Health
   Today’s Solutions
   Personal
   assistants,
   gateways and
   other devices.
   Microprocessors, display
   and stable connectivity
   Microcontrollers and
   sensors
   Wearable monitors,
   mobile applications and
   limited conectivity
   

   View Slide

10. View Slide

11. Alexa Healthcare Skills
    

    View Slide

12. View Slide

13. View Slide

14. {
    "body": {
    "version": "1.0",
    "response": {
    "outputSpeech": {
    "type": "PlainText",
    "text": "Welcome to the Alexa Nursing Skill. Please tell
    me your name"
    },
    …
    

    View Slide

15. View Slide

16. © 2017, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
    AWS IoT Button
    

    View Slide

17. View Slide

18. View Slide

19. View Slide

20. View Slide

21. View Slide

22. View Slide

23. Nordic Thingy:52 and Thingy:91
    

    View Slide

24. View Slide

25. View Slide

26. Security is Job #0
    

    View Slide

27. View Slide

28. View Slide

29. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    

    View Slide

30. View Slide

31. View Slide

32. View Slide

33. View Slide

34. AWS Lam bda
    / R E G I S T E R CH E CK CERT RO B O T S T O
    REG I S T ER
    D E A D L ET T E R
    Q U E U E
    CRE A T E S H A D O W P E RM I S S I O N S
    Am azon SQS
    Messages
    Am azon
    Cl oudWat ch
    L O G G I N G L I F E CY CL E E V E N T
    REG I S T ER
    RO B O T
    Q U E U E
    REA D ER
    AWS
    Lam bda
    AWS
    Lam b da
    A W S I o T
    Am azon API
    Gat ew ay
    I o T Sh ad o w
    I o T Ru l e
    $15k per month to run
    2 IT people to support…”
    Leading Edge Forum 2018
    “… 23M robot vacuum cleaners…
    

    View Slide

35. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    

    View Slide

36. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    Security
    AWS-grade
    security
    Data and
    state sync
    Local
    Device Shadows
    Local
    triggers
    Local
    Message Broker
    AWS IoT Greengrass Features
    Local
    actions
    Local
    AWS Lambda Functions
    

    View Slide

37. Amazon Comprehend Medical
    Entities
    • Medication
    • Medical condition
    • Test, Treatments and Procedures
    • Anatomy
    • Protected Health Information (PHI)
    Relationship Extraction
    • Medication and dosage
    • Test and result
    • Many more
    Entity Traits
    • Negation
    • Diagnosis, Sign or Symptom
    Protected Health Information Identification
    (PHId API)
    Medical Named Entity and Relationship
    Extraction (NERe API)
    

    View Slide

38. Pt is 40yo mother, highschool teacher
    HPI : Sleeping trouble on present dosage of Clonidine.
    Severe Rash on face and leg, slightly itchy
    Meds : Vyvanse 50 mgs po at breakfast daily,
    Clonidine 0.2 mgs -- 1 and 1 / 2 tabs po qhs
    HEENT : Boggy inferior turbinates, No oropharyngeal lesion
    Lungs : clear
    Heart : Regular rhythm
    Skin : Mild erythematous eruption to hairline
    Follow-up as scheduled
    

    View Slide

39. View Slide

40. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
    S U M M I T
    1. Train machine learning model on
    the cloud
    2. Deploy optimized
    model to target devices running
    on Intel or
    NVIDIA hardware
    3. Create and accelerate inference
    applications
    on the edge
    AWS Greengrass ML Inference
    

    View Slide

41. https://github.com/awslabs/Amazon-Pollexy
    

    View Slide

42. View Slide

43. View Slide

44. View Slide

45. Bike safety application
    

    View Slide

46. Food safety application
    

    View Slide

47. View Slide

48. Obrigado!
    Julio Faerman
    @faermanj
    julio.faerman@nubego.io
    

    View Slide