CareFrame: Quick Start Guide

DISCLAIMER: Parts of this documentation have been auto-generated and may contain placeholders. If you find any issues or want to contribute improvements, please submit a pull request to our repository.

Welcome to CareFrame

CareFrame is an integrated platform that streamlines the clinical research process from start to finish. This guide will help you get up and running with the essential features.

📋 CareFrame Installation Guide

System Requirements

• Python 3.9+
• Git
• RAM: Min 16GB (Recommended 32GB+)
• GPU: Recommended 32GB+ VRAM
• Storage: Min 8GB (Recommended 32-64GB)

Installation Steps

1. Installing Python: Download Python 3.9+ from python.org. Check "Add Python to PATH" during installation.
2. Installing Git: Download from git-scm.com for repository access and updates.
3. Run Terminal Commands:

   git clone github.com/CareFrameAI/careframe-research
   cd careframe-research
   python -m venv venv
   source venv/bin/activate  # Windows: venv\Scripts\activate
   pip install -r requirements.txt
   python app_launcher.py
   

4. Local LLM Setup (Optional): Install Ollama (ollama.com) or vLLM (docs.vllm.ai) for privacy and offline operation.
5. CouchDB Database Setup: Install from couchdb.apache.org to store patient data.
6. Configuration: Set up API keys and model settings in the Settings menu after installation.

Setup time: 15-60 minutes depending on your system.

🌟 The Research Workflow

CareFrame organizes research into a connected workflow. Here's how to get started with each phase:

1️⃣ Research Planning

Basic Feature - Essential for all research projects

Start by creating your research plan:

1. Click the Planning icon in the side navigation or home screen
2. Select Create New Plan to open the visual canvas
3. Add your primary research objective:
   • Click the Add Objective button (bullseye icon)
   • Enter your research question
   • Choose the appropriate objective type

4. Add hypotheses to test:
   • Select your objective
   • Click Add Hypothesis (lightbulb icon)
   • Write a clear, testable statement

AI Hypothesis Generation

Advanced Feature - For comprehensive hypothesis development

To generate alternative hypotheses using AI:

1. Select an objective
2. Click Generate Hypotheses (CPU icon)
3. Review and refine the generated options

2️⃣ Literature Search

Basic Feature - Find relevant research papers

1. Navigate to the Literature Search section
2. Enter your search terms in the query box
3. Add filters if needed (publication date, language, etc.)
4. Click Search to retrieve papers
5. Save relevant papers to your collection

Semantic Ranking

Advanced Feature - For in-depth literature review

To rank papers by relevance to your specific hypotheses:

1. Navigate to the Ranking tab
2. Select the hypothesis to use as context
3. Click Rank Papers to sort by relevance
4. Review the top-ranked papers first

3️⃣ Data Collection

Basic Feature - Connect to your data sources

1. Navigate to the Data Sources section
2. Click Add Source to connect to your data
3. Choose from available connection types:
   • CSV/Excel files
   • SQL databases
   • API connections
4. Configure the connection details
5. Preview and confirm the data import

4️⃣ Data Preparation

Basic Feature - Clean and prepare your data

1. With your data source selected, navigate to:
   • Clean for handling missing values and outliers
   • Filter for selecting relevant observations
   • Reshape for transforming data structure
2. Apply transformations as needed
3. Save the prepared dataset for analysis

5️⃣ Statistical Analysis

Basic Feature - Test your hypotheses

1. Navigate to the Model section
2. Select the hypothesis to test
3. Choose the appropriate statistical test
4. Configure test parameters and run the analysis
5. Review the results and visualizations

Assumption Checking

Advanced Feature - For rigorous statistical validity

To verify statistical assumptions:

1. After running a test, go to the Assumptions tab
2. Review automatically detected assumption violations
3. Apply recommended corrections if needed
4. Re-run the analysis with adjustments

6️⃣ Evidence Storage

Basic Feature - Store your findings securely

1. Navigate to the Evidence section
2. Select the hypothesis you've tested
3. Click Push to Blockchain to store the hypothesis
4. Add your evidence:
   • Click Push Model Evidence for statistical results
   • Click Push Literature Evidence for supporting publications

7️⃣ Knowledge Sharing

Advanced Feature - For team collaboration

To share your findings with collaborators:

1. Navigate to the Network section
2. Configure your sharing environment:
   • Host Mode: Set up your computer as a server
   • Client Mode: Connect to another CareFrame instance
3. Select topics to publish or subscribe to
4. Share specific evidence or entire research plans

🤖 Using the AI Assistant

Basic Feature - Get help at any stage

CareFrame includes an AI assistant to guide your research:

1. Click the Agent icon in the top toolbar
2. Ask questions in natural language
3. Request specific guidance on:
   • Statistical test selection
   • Hypothesis formulation
   • Literature search strategies
   • Data cleaning approaches

🔍 Finding Your Way Around

Navigation Tips:

• Side Navigation: Access main modules
• Top Toolbar: Quick access to common tools
• Agent Button: AI assistance
• Settings: Customize your experience

Status Indicators:

• Database: Connection to data storage
• Ollama: LLM service status
• Network: Connection to other instances

👥 Researcher Profiles

CareFrame adapts to different research roles:

• Principal Investigators: Focus on the Research Planning and Evidence modules
• Data Scientists: Concentrate on Data and Statistical Analysis modules
• Clinical Researchers: Prioritize Literature Search and Protocol modules

🎓 Next Steps

After getting familiar with the basics:

• Explore Intermediate Guides for connecting modules
• Review Advanced Documentation for customization
• Check the Workflow Examples for end-to-end research projects

🆘 Getting Help

• Use the Agent for contextual assistance
• Click the ? icon in each module for specific guidance
• Join our community forum for peer support