neuronflow

PS: The code above also implements parallel evaluation (commented code) I reopened the issue, as I agree with you that we need a tutorial to point out these features :)...

@aymuos15 Here some more example code to analyze a single exam: ```python from auxiliary.nifti.io import read_nifti from panoptica import ( InputType, Panoptica_Evaluator, Panoptica_Aggregator, ConnectedComponentsInstanceApproximator, NaiveThresholdMatching, ) from panoptica.utils.segmentation_class import SegmentationClassGroups...

@marwanabb thanks for your interest in blob loss. Yes, it can be used with compound losses both in the global, as well as in the "blob" part. I often did...

Ahh, I remember colleagues at TUM actually used blob loss in a MONAI training loop, but they computed the CCA for each batch again. I prefer to precompute it once,...

> ...with each lesion assigned a unique label. Is this understanding correct? > yes :) > Both options are indeed valid. I was just wondering if the second option could...

Sorry for the late response. Yes, the A6000 you are using seems too new or, in other words, does not have enough backward compatibility to support running these old algorithms....

> 1. Can you not stitch the image together before inputting it into panoptica? no they might just be samples from one bigger entity e.g. cropped slices from one mouse...

seems related to https://github.com/BBillot/SynthSeg/issues/93#issuecomment-2368206469

Why is this a bug and not a feature? This warrants discussion. Currently, the naive underlying design approach was to separate such concerns and put them into the world of...

but is image orientation not a NIfTI-format-specific property? Do numpy arrays have image orientations?