Telestroke - Backend

The Telestroke project aims to create a system to assist doctors (specialist) and first aid personnel (operator) in the assessment of a stroke case gravity (using NIHSS). The system enables the specialist to perform remote reporting, to reduce disease treatment time and improve the quality of the medical supply. This project is my master thesis in Computer Science: Study and development of a remote reporting system.

The project is structured in 3 main components:

• backend: The microservices backend that handles frontend interaction and implements the main application login
• web-frontend: The web application used by the specialist to:
  • implements a WebRTC peer that sends audio to the operator and receives video and audio feed from his/her
  • remotely guide the operator to assess the patient status (by talking to the operator while watching the patient from the operator's feed)
  • guide him/her to perform the medical report and decide the course of action to treat the stroke
• android-frontend: The Android application used by the operator from hands-free wearable device (e.g. Smartglasses):
  • implements a WebRTC peer sends video and audio feed and receives audio from the specialist and receives audio from his/her
  • can display brief information about the current action that the specialist is performing from his/her device

Description

Vert.x microservices architecture to manage authentication, reporting sessions, WebRTC peers and storage.

Configuration

Each microservice can be configured using one or more of the following options:

• resources/service.yml (required)
• resources/service-VERTX_PROFILE.yml
• workingdir/config/*.yml
• k8s configmap named: service-configmap
• k8s secrets named: service-secrets

To see the available configuration for each specific microservice, see the file resources/service.yml inside each gradle module.

Deployment

Jar

gradlew makeDist
java -jar %jar_path%
#or
gradlew run

The makeDist gradle task will build the services as jars inside the folder dist.

Docker

# build images
gradlew buildImage

# deploy
docker-compose up -d

Running ports

Service	Port
gateway	8001
auth-service	8002
registry-service	8003
session-service	8004
peer-service	8005
record-service	8006
frontend	8080
peerjs-server	9000

Known issues

• If you're running Docker Desktop make sure to Expose daemon on tcp://localhost:2375 without TLS. Otherwise gradle buildImage task will fail.

• If you're running Docker Desktop and docker-compose command cannot create volume, use docker volume prune and retry.

• If mongodb skips initialization scripts execution, make sure that the database is empty (prune docker volumes if necessary).

Demo information

The docker-compose is already setup to insert initial data to the database (just for demo purpose). The following table shows the available users:

Username	Password
admin	test
guest	guest