Nathan Molinier comments

Results 35 comments of


                                            Nathan Molinier

Improve `sct_label_vertebrae` using a deep learning-based approach

> ⚠️‎‎‎ **A more recent and actively-maintained version of this code is available in [spinalcordtoolbox/disc-labeling-hourglass](https://github.com/spinalcordtoolbox/disc-labeling-hourglass)**

Improve `sct_label_vertebrae` using a deep learning-based approach

Improve `sct_label_vertebrae` using a deep learning-based approach

Thanks @joshuacwnewton for your feedback ! On my side, I am also doing tests with SPINEPS, I will update you all about what will be the best strategy according to...

Improve `sct_label_vertebrae` using a deep learning-based approach

## My experiments Regarding my work on SPINEPS, I [forked](https://github.com/NathanMolinier/spineps) the repository to: - simplify the CLI to really simple inputs: - "--path-in": corresponding to the niftii image path -...

Improve `sct_label_vertebrae` using a deep learning-based approach

> So, is the idea that only the "label extraction" function is the new part, and the rest of `simple_run.py` is just for convenience/testing? Yes indeed ! The purpose of...

Training dataset identification for sct deep learning based methods

I will have to check for sct_deepseg_sc and sct_label_vertebrae to be fair in the comparison of the methods.

Refactor `sct_deepseg` to store model orientation in model dict

> code like [this one](https://github.com/sct-pipeline/contrast-agnostic-softseg-spinalcord/blob/32ef5c9d541bb779453cca6fc8c25b768bc2c9b1/monai/run_inference_single_image_v2.py#L89) is supposed to take care of this, no? Yes it is indeed the case, however I don't know if nnUNet is performing any reorientations beforehand....

Nathan Molinier

Improve `sct_label_vertebrae` using a deep learning-based approach

Improve `sct_label_vertebrae` using a deep learning-based approach

Improve `sct_label_vertebrae` using a deep learning-based approach

Improve `sct_label_vertebrae` using a deep learning-based approach

Improve `sct_label_vertebrae` using a deep learning-based approach

Training dataset identification for sct deep learning based methods

Refactor `sct_deepseg` to store model orientation in model dict

Refactor `sct_deepseg` to store model orientation in model dict

Sagittal mosaic QC report is tiny for 1.0mm istropic image with 75 slices

Add T2w dorsal rootlets labels