Healthcare Resource Allocation

Big Data Technologies Project

The primary goal of the application is to provide a comprehensive, data-driven view of the Australian healthcare ecosystem, revealing trends and variations in resource utilisation across hospitals and regions.

Designed for government decision-makers, this app enables informed policy-making by providing a clear, real-time snapshot of healthcare performance.

Contents:

1. Project Structure
2. Getting Started
3. Workflow
4. Dashboard Demo
5. Data Sources
6. Authors

1. Project Structure

.
├── LICENSE
├── README.md
├── data
│   ├── Admitted Patients.xlsx
│   ├── Expediture.xlsx
│   ├── Hospital-resources-tables-2022-23.xlsx
│   ├── images
│   │   └── logo.png
│   └── myhosp-mapping-details-extract.xlsx
├── docker-compose.yml
├── dockerfiles
│   ├── Dockerfile
│   ├── Dockerfile2
│   └── Dockerfile3
├── media
│   ├── ABS_logo.jpeg
│   ├── AIHW_logo.png
│   ├── Dashboard budget.gif
│   ├── Dashboard demo.gif
│   ├── Dashboard hospitals.gif
│   ├── Dashboard measures.gif
│   └── logo.png
├── run.sh
└── src
    ├── app
    │   ├── dashboard.py
    │   └── requirements_dashboard.txt
    └── processing
        ├── ETL.py
        ├── jars
        │   └── postgresql-42.7.3.jar
        ├── requirements_spark.txt
        ├── setup.py
        └── utilities
            ├── budget_lm.py
            ├── tables.py
            ├── tools.py
            ├── values.py
            └── values_lm.py

The repository is organized into several key directories and files:

• data/: This folder contains all the relevant datasets in Excel format, along with an images/ subdirectory for logos and other visual assets.

• dockerfiles/: Contains multiple Dockerfiles to support the various environments required by the project.

• media/: Includes media assets such as logos, workflow diagrams, and animated GIFs demonstrating the dashboard features.

• src/: The core source code is located here. It is divided into:

  • app/: Contains the dashboard application script and its dependencies.
  • processing/: ETL scripts, utility functions, and supporting libraries, including JAR files for data processing.

• Root Files:

  • LICENSE: The licensing information for the project.
  • README.md: The main documentation file.
  • docker-compose.yml: Configuration file for Docker Compose to orchestrate the containerized environment.
  • run.sh: A shell script to execute the project setup.

2. Getting Started

1. Clone the Repository:

   git clone https://github.com/SoniaBorsi/Healthcare-Resource-Allocation.git
   cd Healthcare-Resource-Allocation

2. Run the application:

   bash run.sh

3. Wait for the the data to be loaded into the db This usually takes about 1 hour, depending on your internet connection
4. Access the dashboard Got to localhost:8080 and explore all the analytics

Note: It is recommended that the dashboard is run once the data storing process is complete, in line with the current project status. This will be improved in a future update.

3. Workflow

The development of the Australian Efficient Resource Allocation App has been supported by a carefully selected suite of big data technologies, each chosen for its ability to address specific aspects of the system's architecture.

These technologies work seamlessly together to create a robust, scalable application capable of efficiently processing large volumes of health data.

4. Data Sources

Healthcare data utilized in this project is primarily extracted from the Australian Institute of Health and Welfare (AIHW) API, an open and freely accessible resource. The available data includes:

• Hospital Data
• Multi-Level Data
• Geographic Data

In addition, this project integrates healthcare data with supplementary datasets provided by the Australian Bureau of Statistics (ABS) to enrich the analysis and ensure comprehensive insights.

5. Dashboard Demo

This application is designed to facilitate healthcare resource allocation through data-driven insights. Navigate through the various sections using the sidebar to explore different metrics and tools available to you:

• Measures: for detailed metrics for various healthcare measures across different states in Australia

• Hospitals: for a comprehensive analysis of hospitals across different states

• Budget: for a a comprehensive visual analysis of healthcare expenditure data

More details can be found on our paper.

6. Authors

• Borsi Sonia
• Filippo Costamagna