337 - Whole Slide Image segmentation for nuclei using Detectron2 and YOLOv8
ฝัง
- เผยแพร่เมื่อ 10 ก.ย. 2024
- This tutorial builds upon the exercise covered in our previous tutorial ( • 336 - Nuclei segmentat... ), where we trained Detectron2 and YOLOv8 models for Nuclei detection. In this continuation, we will utilize these trained models for segmenting whole slide images with .svs extensions. Additionally, we will demonstrate the use of memory-mapping to efficiently handle large arrays.
Sample svs file used in the tutorial: github.com/cam...
A few large svs images can be downloaded from here: openslide.cs.c...
Code from this tutorial is available here: (Just look for number 337 directory on my GitHub page: github.com/bns...
Direct link here: github.com/bns...
How can we use mIoU instead of mAP for validation of YOLOv8 Segmentation Model ?
Please make a video series of classifications and segmentations using Transformers. Looking forward to explore Transformer based segmentation and classification approaches.
Hello, This is very helpful content. Do you have any idea on how we can make tile predictions in parallel to make whole image prediction faster. I have very big WSI images and sequentially processing would take ~4 days since I have to process 72K tiles in total for one WSI.
Thanks!
Thank you
what is size of those images you segmented?
Thank you for your another video
Can you make a video on Kits23 data set for segmentation using monai
Thank you very much, please can you do brain tumour segmentation using transfer learning... Thank you....
thank you so much,, 😌
You're welcome 😊
I know these videos have gone really really ahead(advanced).But a request from a student/fan. Can u please make a video on haar feature detection. 😊
Tak!
Thank you very much.
@@DigitalSreeni it's so far from the value your videos give to us all, that you thanking me makes me embarrased I didn't give more. Thanks for all your hard work.
Thanks!
Thank you for being very generous.