---
title: MedicalChatBot
emoji: 🔥
colorFrom: yellow
colorTo: blue
sdk: gradio
sdk_version: 6.12.0
app_file: app.py
pinned: false
---

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

# Patient/Doctor Medication Management Chatbot
   
   ## Domain
   Medication Management
   
   ## Setup Instructions
   ### Prerequisites
   - Python 3.14.4+
   - Ollama
   - Gradio
   
   ### Installation
1. Install Ollama 
   ```bash
    # Mac:
     brew install ollama 
    
    # Windows: Download from https://ollama.ai 
    
    # Linux:
    curl -fsSL https://ollama.com/install.sh | sh 
   ```

1. Download llama3.2:3b model:
   ```bash
   ollama pull llama3.2:3b
   ```
1. Install Python packages 
   ```bash
      pip install ollama gradio
   ```
   
   ### Running

    ```bash
    python your_file.py
    ```
   
   ## Features
  
  - Feature 1: Answer patient questions about medications and remembers history.

  - Feature 2: Provide doctors with summaries of patient inquiries to help with medication management.

  - Feature 3: Suggests prompts depending on the mode selected (different prompts for doctors and patients).

  - Feature 4: Export conversation history for record-keeping or further analysis.

    
   
   ## Technical Details
   - Model: llama3.2:3b
   - Framework: Ollama
   
   ## Demo
   [Link to video OR screenshots]
   
   ## Known Limitations

   Currently, both patient and doctor modes are on a shared user interface. This serves as a prototype to easily see how patients and healthcare professionals can leverage this tool to understand a patient's lived experience with their medication.

   Additionally, the long-term memory of this tool is minimal. After several prompts, the chatbot will begin to respond with blanks because the context is overflowing.
   
   ## Future Improvements

   With more time, I would build out this interface to work as two independent applications.
   One for patients, and one for Healthcare professionals.

   Use a more dynamic way of keeping the context of a patient's prompts, over multiple days, and without causing issues for the model. Local caching of patient's prompts in summarization form may allow a more concise version to be passed in as context. This would mean the doctor would get a more accurate idea of the patient's questions over a longer period of time.