We are pleased to be able to share with you the first part of our prototype that our developer Joan has been working on so tirelessly.
Joan has processed the images based on open source data we are using and they are ready to feed into the detection algorithm.
DICOM images are being used as they are they provide a lot of metadata, such as pixel size, radio density, etc
After applying an algorithm to make the calculations and fill in the gaps, we can see an example of a slice of one of the patients:
This information will be later used for lung segmentation.
We create some plots of the entire CT scans. This allows us to calculate, for example, the bone structure and plot it:
We take this approach and apply some more complex processing techniques in order to segment the lungs.
And then add the structures inside the lungs to our model.
We are now ready to feed the processed images to our model.