sct-pipeline/contrast-agnostic-softseg-spinalcord

Issues

• tag and release should be the same

  #124 opened 7 days ago by jcohenadad
  0

• Incorrect segmentation on a thoracic healthy patient

  #122 opened 14 days ago by po09i
  10

• Release strategy for models

  #123 opened 14 days ago by jcohenadad
  1

• Datasets to add for the next version of the model

  #116 opened 3 months ago by naga-karthik
  2

• Contrast-agnostic model on a lumbar SCI patient outputs incorrect segmentation

  #118 opened 2 months ago by naga-karthik
  10

• open PR to merge code to generate nice figure

  #121 opened 16 days ago by jcohenadad
  0

• `nnunet_mednext` is not listed in `monai/requirements.txt`

  #114 opened 21 days ago by valosekj
  2

• Remove `datasplit_sci-zurich_seed50.yaml` from `v2.4` tag

  #120 opened 23 days ago by valosekj
  1

• Increase probability of existing transforms

  #119 opened a month ago by naga-karthik
  0

• Segmentation doesn't go as high as C1 disc label

  #117 opened 2 months ago by sandrinebedard
  3

• Memory issue with training on Monai with large datasets

  #59 opened a year ago by louisfb01
  20

• Shifted predictions

  #113 opened 4 months ago by jcohenadad
  9

• Consider binarizing the ground truth after averaging across contrasts, before training a softseg model

  #99 opened 10 months ago by jcohenadad
  4

• Upload QC of test set

  #115 opened 4 months ago by jcohenadad
  2

• Re-run preprocessing to include combinations of contrasts

  #112 opened 4 months ago by sandrinebedard
  1

• Training SynthSeg on spine-generic data

  #111 opened 5 months ago by naga-karthik
  6

• Inconsistent list of subject used for training

  #109 opened 5 months ago by jcohenadad
  7

• Add post-processing code to further improve prediction

  #73 opened 5 months ago by jcohenadad
  1

• run_inference_single_image.py v1 vs v2

  #106 opened 5 months ago by jcohenadad
  3

• Consider moving this repository to ivadomed

  #110 opened 5 months ago by jcohenadad
  0

• Testing contrast-agnostic model on GRE magnitude data

  #108 opened 5 months ago by jcohenadad
  1

• QC compression results between monai and deepseg_2d models

  #107 opened 5 months ago by naga-karthik
  1

• Add reorientation to RPI also for train/val transforms

  #103 opened 6 months ago by valosekj
  3

• Naming convention for adding new contrasts to the training set

  #102 opened 8 months ago by naga-karthik
  12

• Create scripts to reproduce results

  #101 opened 9 months ago by jcohenadad
  1

• Strategy to re-use existing binary ground truth for enriching contrast-agnostic model

  #84 opened 8 months ago by jcohenadad
  26

• Fix shape mismatch with different input sized UnetR training script

  #54 opened 8 months ago by louisfb01
  2

• What threshold to use on model prediction?

  #98 opened 8 months ago by jcohenadad
  2

• Make CLI callable from anywhere

  #94 opened a year ago by jcohenadad
  0

• Doc points to non-existing file

  #92 opened a year ago by jcohenadad
  1

• Implement strategy for assessing the quality of the model during lifelong training

  #87 opened a year ago by jcohenadad
  2

• Extend model to also segment the CSF

  #89 opened a year ago by jcohenadad
  0

• Cleanup repos

  #62 opened a year ago by jcohenadad
  2

• Re-train `contrast-agnostic` model with `EPI` data

  #83 opened a year ago by rohanbanerjee
  1

• Slow repo cloning

  #86 opened a year ago by valosekj
  6

• Running into errors while using `--device gpu` while running inference using MONAI

  #75 opened a year ago by rohanbanerjee
  3

• Disc labels files created by `process_data.sh` are empty for `sub-brnoUhb` and `sub-tokyo750w` sites

  #63 opened a year ago by valosekj
  2

• Errors while using the `monai` training script for the EPI data

  #72 opened a year ago by rohanbanerjee
  4

• Test contrast-agnostic model in Marseille 7T T2S data

  #74 opened a year ago by Nilser3
  0

• Update README on how to run training

  #49 opened a year ago by sandrinebedard
  1

• Prediction clipping sets background values to 0.5 instead of 0

  #71 opened a year ago by valosekj
  2

• One outlier subject in the softseg CSA of T1w contrast

  #66 opened a year ago by naga-karthik
  7

• Compute CSA prediction error

  #67 opened a year ago by sandrinebedard
  5

• UnetR training script

  #52 opened a year ago by louisfb01
  3

• Work on a test notebook to identify the problem step by step and visualize each step of the pre-processing.

  #56 opened a year ago by louisfb01
  1

• Try with much smaller patch sizes (based on nnunet recommended size @ebadrian ?).

  #55 opened a year ago by louisfb01
  1

• Analysis for the paper

  #68 opened a year ago by sandrinebedard
  3

• When to center crop images at test-time

  #70 opened a year ago by naga-karthik
  1

• Augmentation with intensity scaling

  #69 opened a year ago by naga-karthik
  0

• Train on with other seeds

  #48 opened 2 years ago by sandrinebedard
  3