kmit-cadsc-web

A showcase of the work (being) done in the KMIT CADSC project.

Running Instructions

To view the website, simply run python3 server.py and then go to http://localhost:5010.

• Before running the mammography pipeline download the weights of vgg16(the base layer) here and the weights of our(network) here and paste it in pipelines/mammography/model.

For colleagues working on the project

Adding your task to the website

Let us say you need to add a new task to the website, called "Pizza Grading".

Creating a webpage for your task

• Go to templates/partials/header.html and inspecting the line above "Add new task button here", add a similar line for your task. For Pizza Grading, this line would look like:

<a class="item" href="/tasks/pizza-grading">Pizza Grading</a>

• Create a new HTML file in templates named as your task in snake case ("pizza_grading.html").
• Copy the contents of templates/epithelium_segmentation.html, paste it into your newly created HTML file and edit it as you see fit. The parts that you might want to edit are highlighted with "Might want to edit this".
• Now open up the main page, click on Menu, and you should be able to access your new page!

Writing the image processing code for the 'Test it out!' section

• Create a new folder named as your task in snake case ("Pizza Grading" => "pizza_grading").
• Here, create a file __init__.py. This is where you write all the image processing code. Write a function that takes in a file name, and generates the output image.
• When the 'Go!' button is pressed, the function process_image in server.py is called, so you may need to edit process_image to work with the above code.
• If everything goes well, taking in input images in the form and hitting 'Go!' should redirect you to the output image.

Finishing touches

• Add sample inputs of your task to the sample_inputs directory.

Updating the repo

• Just create a pull request against master.