SupportPrim CDSS Backend Architecture

Welcome to the repository for SupportPrim CDSS (Clinical Decision Support System), solely developed by Amar Jaiswal as part of his PhD research:

Leveraging Knowledge and Data-Driven Approaches in Case-Based Reasoning to Aid Physiotherapists Addressing Non-Specific Musculoskeletal Disorders

---

1. Introduction

SupportPrim CDSS assists physiotherapists in making informed decisions for non-specific musculoskeletal disorders (MSDs). It uses Case-Based Reasoning (CBR), integrating both knowledge-driven (expert rules) and data-driven (machine learning) approaches to match patient data with previously successful treatments. This approach streamlines clinical workflows, supports evidence-based practice, and offers personalized treatment recommendations.

2. Key Features

1. Hybrid Case Retrieval
   • Merges data-driven machine learning techniques with expert-encoded rules.
2. Adaptive Learning
   • Updates its recommendation engine with outcomes from newly treated cases.
3. Comprehensive CBR Workflow
   • Follows the Retrieve, Reuse, Revise, Retain cycle for tailored recommendations.
4. Interoperable Microservices
   • RESTful APIs ensure easy integration with Electronic Health Records (EHRs) and third-party systems.
5. Scalable Design
   • Supports distributed deployments and manages large volumes of patient data.

---

3. Deployment Structure

supportprim-cdss/
├── README.md
├── jars/
│   ├── cbr-interface/
│   │   ├── application.properties
│   │   └── cbr-interface-0.1-SNAPSHOT.jar
│   ├── data-management-0.1-SNAPSHOT.jar
│   ├── discovery-server-0.1-SNAPSHOT.jar
│   ├── elastic-interface-0.1-SNAPSHOT.jar
│   ├── infopad-interface-0.1-SNAPSHOT.jar
│   ├── lca-calculator-0.1-SNAPSHOT.jar
│   ├── mapping-0.1-SNAPSHOT.jar
│   ├── myCBR-3.3-SNAPSHOT.jar
│   ├── mycbr-rest-2.0.jar
│   └── transformer-0.1-SNAPSHOT.jar
├── mycbr-projects/
│   └── supportprim_cbr_latest.prj
├── resources/
│   └── infopad_to_ism_transformer.xslt
├── start-backend.sh
├── stop-backend.sh
└── ...

• jars/: Java-based services and dependencies.
• mycbr-projects/: Files for the myCBR engine.
• resources/: Configuration or transformation files.
• start-backend.sh, stop-backend.sh: Scripts to start and stop microservices.

4. Architecture Overview

1. Discovery Server
   • Registers and discovers available microservices.
2. CBR Interface
   • Connects to the myCBR engine for case retrieval and adaptation.
3. Data Management Module
   • Ingests and stores patient data from various sources.
4. Infopad Interface
   • Transforms and integrates external healthcare data (e.g., Infopad APIs).
5. Transformer & Mapping
   • Ensures consistency in data formats (XML, JSON, etc.).
6. Elastic Interface (Optional)
   • Supports advanced search and analytics with Elasticsearch.
7. UI Integration
   • A front-end (not included in this repo) calls these services via REST.

For a visual representation, see the supportprim-backend-subsystems_v3.pdf included in this repository.

5. Deployment Scripts

• start-backend.sh : Launches microservices in the required sequence.
• stop-backend.sh : Gracefully halts all running services.

6. License

This software is made available “as is,” one must seek permission from SupportPrim pincipal investigator (PI) Ingebrigt Meisingset having email-id as [email protected]. Users must recognize the intellectual efforts of Amar Jaiswal and the academic goals of his PhD research when employing or modifying this code.