Files

.
├── models            # skin lesion diagnosis models
├── datagen.py        # uses dataset.json to generate patient data
├── dataset.json      # stores a list of names, addresses, etc.
├── flaskaimar.db     # the actual database file (.db = database)
├── image_util.py     # handles patient images and skin diagnosis
├── patient_util.py   # patient-related functions (registration, queuing, etc.)
└── routes.py         # handles incoming web requests, wrapper for all other functions

Download

Make sure you have Python 3.7 installed. When installing, check the "Add Python to PATH" option.

On Mac, or Windows (if you have Git for Windows), run:

git clone https://github.com/UMD-AIMAR/desk-server.git

Or go to the top, click 'Clone or download', and click 'Download ZIP'.

Open a Terminal and run:

python -m pip install -r requirements.txt

Running

Open a Terminal/Command Prompt, navigate to the folder 'desk-server', and run:

python routes.py

When prompted for the Desktop IP, type "localhost" (without the quotes).

Then, go to http://localhost:5000 in your browser. You should see a page with the text "AIMAR Homepage!" That's it! You're now running a Flask server on your own computer.

Description

AIMAR's architecture requires a central patient database and diagnosis system. The system runs on its own server (independent of any TurtleBot AIMAR units). Bots interact with the server by sending requests, e.g.

• "Give me patient X's info"
• "Update patient Y's zip code to 12345"
• "Run analysis on this skin lesion image"

From the bot, executing a diagnosis/database function (e.g. diagnosing a skin lesion) can be done with the Python requests library. An example code snippet would look like:

skin_diagnosis_response = requests.post("https://desktop_ip/skin", data=skin_image_data)

Running a command like this in Python will send a request to the internet address 'desktop_ip'. desk-server receives this request and handles it based on the endpoint specified (in this case, /skin). routes.py is responsible for handling different endpoints - all the processing and logic is done in the other Python files.