Python Image Segmentation Tutorial (2022)

This is a tutorial about non-AI based methods to segment images in python. Methods are state of the art.

Code:

Instagram:

Discord:

Paper mentioned earlier in the video:

[isoljator]

This was SUPER interesting! Thank you for methodically going through each step in the process, while providing a very clear and detailed account. Your whole channel is pure gold 🏆

[robertpretorius]

Awesome video! Thanks for showing the plotly thing at the end. I've been banging my head against the wall for a week trying to visualize MRI data in plotly without success.

[carlogalicia]

Hey man, really appreciate what you’re doing. As a grad student myself, it’s really admirable that you teach us this stuff. Best wishes at school and life.

[sin-leqi-unninni]

Your medical physics videos are the best stuff ever

[-TheBugLord]

I've never worked with high level stuff like this, but I really enjoy your walkthroughs, because it helps me understand the thought process of performing complicated tasks such as this. I hope to do something this cool someday.

[tiddlywinks]

Thank you so much for these videos. I have started getting more involved in CT scans, and these videos have been very inspiring.

[idiosinkrazijske.rutine]

I like it so much when people take care about reproducibility of their research especially when it comes to computational science. Code should be supplementary material along with the data, and there shouldn't be "unnecessary hurdles on a way to reproducibility" (a citation from a Nature article as I recall). We have a technology for this like online repositories and code should also get doi numbers.
Thanks for this video!

[gaspi]

This video is amazing, just as libraries for segmentation in python, skimage is OP.

[manasmitapanda]

Learning python and libraries from your channel is fun 😇

[guilhermedelimaschwaikartt]

Hi, my jaw dropped on thie masterpiece, but I have one question:
In the import file, you used a CT_scan.npy, wich means, and correct me if I'm wrong, that you transformed 200 layers of 1 exam that was .dcm in one single archive that is a group of arays .npy.
Do you have any tutorial on how to do that or some link to how can I learn the way you did it? It would help a lot! Thanks!

[poreh]

This is exactly what I have wanted,
I cannot thank you more man!

[alonalon]

Where can I get the '../Data/CT_scan.npy' file that you're using in cell 2? @Mr. P Solver

[domillima]

Excellent video. In my case, I have CT DICOM files that I’ve segmented using 3D slicer. I read those files in with sitk and need to work on processing them with labels the way you did here so that I can multiple the image x the labels and remove background nose before training into my model. Would love more videos along this line!

[enreas]

Amazing video, thanks a lot. I'm trying to follow the steps, but when trying to download the file CT_scan.npy GitHub says the repository is over its data quota. Is there any other way of getting that file?

[Mate-Shorts]

wow you're my savior, i need it ... thanks!

[baobao]

I couldn't find the "CT_scan.npy" file in the Data directory in your Github repo. Btw, do you have any tutorial on how to process the medical imaging files (dcm, MRI, ...)

[vladiS]

FANTASTIC!!!! Love the video!!!! Pls do more of these videos!!!! Thank you!!! :)

[najninmansuri]

Nice description 👍🏻
If i will take several images of lungs and want to integrate them and make one image than how I can integrate that several images?

[mikekertser]

Fantastic!
And definitely cheaper than Image Processing Toolbox in MATLAB ;)

[theantonlulz]

Fuck me silly this is a good quality video. The concrete example really helped me wrap my head around all the material.