yascoma / medicalimagevolume

Summary

This repository is a web application to handle medical images in dcm (dicom) format with the purpose of calculating the volume of image segments that have a higher contrast. This application contains the api with the function to receive the uploaded image and return the volume, and the front end web application that contains a form to drag and drop the image.

Tech stack

• Backend: FastApi
• Frontend: Nuxt 3

Configuring the backend api

• cd api-backend
• docker-compose up --build (in case there is no image yet)
• docker-compose up

Testing the api

• docker exec -it api-backend_web_1 /bin/bash
• pytest

Configuring the frontend web application

• cd webapp-frontend
• docker-compose up --build (in case there is no image yet)
• docker-compose up

Access endpoints

• Backend: http://127.0.0.1:8000
• Frontend: http://127.0.0.1:3002

Future work

This implementation only selects the regions of the image considering a single threshold value, it may lead to find false positives if health regions of the pixels have high intensity values. A better solution would involve applying convolution windows as filters and then apply the volume calculation.