DHMAS is a distributed system that monitors patients' health in real-time and issues alerts based on collected data.

DHMAS system consists of five components:

1. Patient Data Collection Service (PDCS)

• Uses MQTT clients and Mosquitto broker.
• Receives data from simulated devices.
• Forwards the messages to the HAS module.
• The MQTT folder contains a data generator to simulate real-time working medical devices, one for each recorded property:
  • Holter: generates random heart_rate every 15 seconds for 10 patients.
  • Oximeter: generates random SpO2 every 15 seconds for 10 patients.
  • Respirometer: generates random respiratory_rate every 15 seconds for 10 patients.
  • Thermometer: generates random temperature every 15 seconds for 10 patients.

2. Health Analysis Service (HAS)

• Serves as a Consumer of one Kafka queue and receives data from PDCS.
• Ensures a smooth connection between PDCS and HAS.
• Evaluates the received data, adds attributes, and writes to the database using MyBatis.
• Combines information from two tables to generate an alert message.
• Acts as a Producer for another Kafka queue and sends the newly assembled alert message to the ANS service.

3. Alerting and Notification Service (ANS)

• Receives data from the HAS module using Kafka.
• Sends emails to those altered patients.
• Uses MailHog to monitor sent emails.
• Stores received data and sends message information to the database.
• Exposes an API for sending altering emails.

4. Patient Management Service (PMS)

• Handles CRUD Operations.
• Acts as an intermediary between the tables and the information.
• Utilizes Get and Post requests.
• Implements security features as authentication measures in Spring Boot.
• Creates login credentials.

5. Real-time Patient Monitoring Dashboard (PMD)

• Summarizes the total number of different categories.
• Displays all patients’ personal information as a table.
• Adds filters to show what users want to focus on.
• Includes a component to insert new patients into the database.
• Counts and visualizes patients’ information as charts.

• Under the project folder (dhmas-newest), run docker-compose up
• You may have to wait for 5 minutes to allow all services being ready
• You will see 14 containers, each built from one folder (with the same name):
  • mosquitto
    • This container is acting as a broker with sensors and PDCS.
  • oximeter, thermometer, respirometer, holter.
    • Those 4 containers are used to simulate sensors to generate user data.
  • pdcs
    • This container is the Patient Data Collection Service.
    • Connects 4 sensors through mosquitto.
    • Sends data to HAS using Kafka. The topic is collection-analysis.
  • has
    • This container is the Health Analysis Service.
    • Receives data from PDCS using Kafka. The topic is collection-analysis.
    • Sends data to ANS using Kafka. The topic is analysis-alert.
    • Stores received data into database using MyBatis.
  • ans
    • This container is the Alerting and Notification Service.
    • Receives data from HAS using Kafka. The topic is analysis-alert.
    • Stores received data into database using JPA.
  • pms
    • This container is the Patient Management Service.
    • Performs CRUD on two tables.
    • Modifies the database using JPA.
  • frontend
    • This container is the deployment of the front end React app.
    • Displays the information of the general_info table.
    • Shows dynamic alerting messages based on the status_localis table.
    • Interacts with PMS and ANS using the Restful API.
  • database
    • This container is the shared PostgreSql database.
    • Initializes 3 tables by init.sql script.
  • mailhog
    • This container is the fake SMTP server.
    • Used to show the sent emails.
    • Used by the ANS service
  • kafka
    • This container is the Kafka service used as a message queue between PDCS, HAS, and ANS.
    • The docker-compose has KAFKA_CREATE_TOPICS: 'collection-analysis:1:1,analysis-alert:1:1 settings so it will have two topics ready after built.
• Open localhost:3000 to check the frontend

• To start, run minikube start, which downloads the Minikube image into Docker.
• Under the project folder run ./deploy.sh. If it says lacking permission, please run chmod +x deploy.sh.
• Wait for all pods to get ready. You can check the pods' status using kubectl get pods. The pods may show errors; just wait since Kubernetes will restart pods automatically.
• After all pods are running, run minikube service react-app --url. You will see a URL after running.
• Open the URL in the browser, and you will see the same website as in the Docker Version.