dzuluaga / doctor-digital

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

doctor-digital

Overview

We have an accelerator called Apigee HealthAPIx software solution which assists healthcare providers in accelerating the development and delivery of digital services based on FHIR (Fast Healthcare Interoperability Resources) APIs.

Apigee HealthAPIx is built on the Apigee Edge API management platform, and features FHIR APIs and a healthcare developer portal to help hospitals meet the demand for data interoperability, deliver patient-centric healthcare, and move faster to the digital world.

http://fhirtest.uhn.ca/ is a publicly accessible FHIR server We are using the DSTU2 FHIR server for this demo - http://fhirtest.uhn.ca/baseDstu2. It is accessible using the HAPI client implementation.

We have interfaced some of these APIs(appointment booking) using dialogflow.

Components

Dialog flow agent -

These are modules which convert user requests to actionable data

Apigee Edge -

  • apiai-connector - webhook for the dialog flow agent. This is where all the actionable data is used to form an RESTful request for the provider.
  • fhir-api - This is a part of the flame solution which is built as an accelerator. This is the northbound part of the proxy to which the calls are made by the apiai-connector.
  • fhir-connector - This is also part of the flame solution. This is the southbound part of the proxy which connects to the publically accessible FHIR server.
  • auth-b2b - This is also a part of the flame solution. This is a Business2business authorization module for the FHIR APIs exposed as a part of the flame solution.

Installation Instructions

Install edge componenets

Pre-requisites

Apigee edge account (https://enterprise.apigee.com)
node.js and npm
curl -o- https://raw.githubusercontent.com/creationix/nvm/v0.33.8/install.sh | bash 
export NVM_DIR="$HOME/.nvm"
[ -s "$NVM_DIR/nvm.sh" ] && \. "$NVM_DIR/nvm.sh" # This loads nvm
[ -s "$NVM_DIR/bash_completion" ] && \. "$NVM_DIR/bash_completion"
nvm install --lts
Clone the repository
git clone https://github.com/rohan-m/doctor-digital.git
Enter edge directory
cd doctor-digital/edge/
Install gulp
npm install --global gulp-cli

If you run into permission issues, try sudo npm install --global gulp-cli

Pull node modules
npm install
Run the deploy command
gulp deploy --target_server_host fhirtest.uhn.ca --target_server_port 80 --target_server_basepath /baseDstu2

This will interactively prompt you for following details, and will then create / deploy all relevants bundles and artifacts and will provision the Doctor Digital Sandbox on your own Org.

  • Edge Org name
  • Edge Username
  • Edge Password
  • Edge Env for deployment

Install dialogflow components

Pre-requisites
Dialogflow account(https://dialogflow.com)
  • Click on Go to console
  • Sign-in with your google account
Create a new agent

Before uploading the zip, please create a new agent, which is inturn assigned to a google project(default is to create a new one)

Import zip
  • Go to settings of the agent
  • Click on Export and Import tab
  • Select the Import from zip
  • Include the zip provided in this repository dialogflow/doctor-digital-health.zip
Modify fulfilment
  • Click on Fulfilment
  • Change the Webhook url to include the Edge organisation, environment and the API key of the api ai connector app https://{org}-{env}.apigee.net/api/ai/connector/fhir?apikey={apiAiApp_consumerkey}
  • Developer app apiAiApp got created in your org by the previous step, copy the consumer key
Add Integrations
  • Click on Integrations
  • Click on Google Assistant Integration Settings
  • You will find TEST and MANAGE ASSISTANT APP, click on TEST to start testing right away. Find the [optional] section here to add more details to the application like name, voice, other innvocations,
  • You will visit Actions on Google simulator page
  • Please make sure that testing on device is enabled, which is the found at the first icon in the top right corner of the page
  • [Optional start] ----
  • Click on MANAGE ASSISTANT APP to configure the assistant
  • You will visit Actions on Google information page.
  • You find App Information. Click on it and click Edit.
  • Click on Assistant app name to change its name.
  • Click on Details, you will find introduction, app voice, descriptions. You can fill them(its optional). In this same section you will find same invocations, ignore the existing ones, you can add upto 5 invocations(like Connect to Digital Health or Connect to Doctor Digital). Additionally you can also add images and other minute details.
  • Save the changes
  • Try Test Draft or Back to Dialogflow page click TEST
  • [Optional end] ----

Usage Steps

Try the following on the actions simulator or your Google assistant app on your phone

  • Start with saying Connect to Digital Health or Talk to Digital health or any other invocations that may be set
  • It will ask for your phone number - 669 5858 is the default number to use, you can add a patient with a new number and a name and you can specify the same(to create a patient record and clinic location record find below optional section)
  • Once you have provide your phone details there are couple of things you can do Like
    • Get nearby clinic
    • Show my appointments - then delete my appointments
    • Book a new appointment -works only when you have got the clinic info
    • Get wait time at the clinic -works only when you have got the clinic info
  • The assistant will end the conversation after a successful booking or deletion of an appointment, you may have to reconnect to list the appointment
  • There is a slight delay in listing the appointment as the data in the FHIR server is reflect after a few seconds.
  • [Optional start] ----
  • Please note
  • While creating a patient record, make sure that the phone number you add is unique among your peers and is pronounceable easily for demo purpose :)
curl -X POST \
  http://rmahalank1-eval-test.apigee.net/digital/health/admin/Patient \
  -H 'cache-control: no-cache' \
  -H 'content-type: application/json' \
  -d '{
    "given_name": "Rohan",
    "family_name": "M",
    "phone_number": "6694747"
  }'
  • Please note
  • Fill in complete details to the accurate, especially the latitude and longitude as they are used to determine the clinics near to you.
curl -X POST \
  http://rmahalank1-eval-test.apigee.net/digital/health/admin/Location \
  -H 'cache-control: no-cache' \
  -H 'content-type: application/json' \
  -d '{
    "hospital_name":"Digital University Medical Center",
    "addr_line":"Central Park",
    "city":"Bangalore",
    "state":"KA",
    "postal_code":"560033",
    "country":"IN",
    "lat":"12.994261",
    "long":"77.660916"
  }'

  • [Optional end] ----

About

License:Apache License 2.0


Languages

Language:JavaScript 100.0%