Comments (7)
I checked the masks with multiple clusters, which were probably due to misclicks during the manual segmentation process. They have been cleaned and uploaded. The T2 lesion masks should only contain a single, correct cluster now.
I have also checked patient 199, I will correct the segmentation shortly in consensus with the other senior reader and upload the updated masks. I will let you know as soon as it is done.
from prostatex_masks.
Thank you for letting me know. I will check these masks again in the next days, I will upload corrected files as soon as possible. For patient 199, I will look again at the masks and original coordinates (corresponding 3x3 masks have now been included in the repository if you are interested) to assess how best to proceed.
from prostatex_masks.
Re patient 199, the two masks seem to be in right place based on the coordinates. They are just overlapping slightly since they are both large and in the central gland. Thanks Renato.
from prostatex_masks.
Sorry for the wait. Today I was finally able to work together with the other radiologist involved in the repository and we have redone the annotations for the T2 images of case 199. I believe they have been improved overall and the overlap should also be fixed. I would appreciate it if you could also confirm the masks are now correct.
from prostatex_masks.
No problem. Thanks Renato. The masks for case 199 look good (to my untrained eye). The overlap has also gone as desired. However, there is still a point of contact between the masks on one of the slices (slice 9 I believe). I think the point of contact should be removed, else the masks are essentially describing one lesion. What are your thoughts?
from prostatex_masks.
I have checked again, on slice 9 the masks become contiguous but do not overlap. The issue is that this lesion could actually also be considered one large lesion rather than two separate ones. In the T2w images it is very challenging to actually separate them, while it is clearer on ADC maps. However, in the TZ the T2w are the reference standard, while DWI/ADC has an ancillary role. Therefore, lesion extension is better reflected by T2w images. As the PROSTATEx dataset considers these 2 separate lesions and the biopsy cores are reported quite fairly apart according to lesion coordinates, we have kept this classification. We believe the current mask to be a good compromise to actually include all suspicious tissue according to PI-RADS criteria while also adhering to PROSTATEx labels. I would not edit it further, honestly.
from prostatex_masks.
Thanks Renato. Understood. I'm happy for this issue to be closed. Thank you and your team once again for all the efforts.
from prostatex_masks.
Related Issues (19)
- [Q] ProstateX-0025 & ProstateX-0113 HOT 4
- Unmatched size of prostate masks HOT 4
- ADC mask slice order is inverted/flipped according to the ADC dicom slices HOT 1
- Varying T2WI sequence used for drawing prostate/zonal masks and lesion masks HOT 1
- Request for radiomic feature extraction HOT 2
- Submit to TCIA? HOT 5
- download link of image nii files HOT 5
- Can a list of training and test sets be provided? HOT 1
- Can Pirads score of each findings be provided? HOT 1
- Question about the dataset
- A question about the marked lesion mask HOT 1
- Errors in several zone segmentation masks HOT 4
- Multi-parametric lesion segmentation HOT 3
- Further details about contouring HOT 2
- Radiologist scores for annotated lesions HOT 12
- How to cite? HOT 6
- number of slices in masks is not always the same as number of slices in the scan HOT 5
- There is something wrong about the ProstateX-0142 Finding 3 geometry HOT 11
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from prostatex_masks.