A recurrent network for segmenting the thrombus on brain MRI in patients with hyper-acute ischemic stroke

Sofia Vargas Ibarra, Vincent Martin VIGNERON, Sonia Garcia Salicetti, Hichem Maaref, Jonathan Kobold, Nicolas Chausson, Yann Lhermitte, Didier Smadja
Proceedings of The 7nd International Conference on Medical Imaging with Deep Learning, PMLR 250:657-671, 2024.

Abstract

In the stroke workflow, timely decision-making is crucial. Identifying, localizing, and measuring occlusive arterial thrombi during initial imaging is a critical step that triggers the choice of therapeutic treatment for optimizing vascular re-canalization. We present a recurrent model that segments the thrombus in patients suffering from a hyper-acute stroke. A cross-attention module is defined to merge the diffusion and susceptibility-weighted modalities available in Magnetic Resonance Imaging (MRI), which are fed to a modified version of the Convolutional Long-Short-Term Memory (CLSTM) model. It detects almost all the thrombi with a Dice higher than 0.6. The lesion segmentation prediction reduces the false positives to almost zero and the performance is comparable between distal and proximal occlusions.

Cite this Paper

BibTeX
@InProceedings{pmlr-v250-ibarra24a, title = {A recurrent network for segmenting the thrombus on brain MRI in patients with hyper-acute ischemic stroke}, author = {Ibarra, Sofia Vargas and VIGNERON, Vincent Martin and Salicetti, Sonia Garcia and Maaref, Hichem and Kobold, Jonathan and Chausson, Nicolas and Lhermitte, Yann and Smadja, Didier}, booktitle = {Proceedings of The 7nd International Conference on Medical Imaging with Deep Learning}, pages = {657--671}, year = {2024}, editor = {Burgos, Ninon and Petitjean, Caroline and Vakalopoulou, Maria and Christodoulidis, Stergios and Coupe, Pierrick and Delingette, Hervé and Lartizien, Carole and Mateus, Diana}, volume = {250}, series = {Proceedings of Machine Learning Research}, month = {03--05 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v250/main/assets/ibarra24a/ibarra24a.pdf}, url = {https://proceedings.mlr.press/v250/ibarra24a.html}, abstract = {In the stroke workflow, timely decision-making is crucial. Identifying, localizing, and measuring occlusive arterial thrombi during initial imaging is a critical step that triggers the choice of therapeutic treatment for optimizing vascular re-canalization. We present a recurrent model that segments the thrombus in patients suffering from a hyper-acute stroke. A cross-attention module is defined to merge the diffusion and susceptibility-weighted modalities available in Magnetic Resonance Imaging (MRI), which are fed to a modified version of the Convolutional Long-Short-Term Memory (CLSTM) model. It detects almost all the thrombi with a Dice higher than 0.6. The lesion segmentation prediction reduces the false positives to almost zero and the performance is comparable between distal and proximal occlusions.} }
Endnote
%0 Conference Paper %T A recurrent network for segmenting the thrombus on brain MRI in patients with hyper-acute ischemic stroke %A Sofia Vargas Ibarra %A Vincent Martin VIGNERON %A Sonia Garcia Salicetti %A Hichem Maaref %A Jonathan Kobold %A Nicolas Chausson %A Yann Lhermitte %A Didier Smadja %B Proceedings of The 7nd International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2024 %E Ninon Burgos %E Caroline Petitjean %E Maria Vakalopoulou %E Stergios Christodoulidis %E Pierrick Coupe %E Hervé Delingette %E Carole Lartizien %E Diana Mateus %F pmlr-v250-ibarra24a %I PMLR %P 657--671 %U https://proceedings.mlr.press/v250/ibarra24a.html %V 250 %X In the stroke workflow, timely decision-making is crucial. Identifying, localizing, and measuring occlusive arterial thrombi during initial imaging is a critical step that triggers the choice of therapeutic treatment for optimizing vascular re-canalization. We present a recurrent model that segments the thrombus in patients suffering from a hyper-acute stroke. A cross-attention module is defined to merge the diffusion and susceptibility-weighted modalities available in Magnetic Resonance Imaging (MRI), which are fed to a modified version of the Convolutional Long-Short-Term Memory (CLSTM) model. It detects almost all the thrombi with a Dice higher than 0.6. The lesion segmentation prediction reduces the false positives to almost zero and the performance is comparable between distal and proximal occlusions.
APA
Ibarra, S.V., VIGNERON, V.M., Salicetti, S.G., Maaref, H., Kobold, J., Chausson, N., Lhermitte, Y. & Smadja, D.. (2024). A recurrent network for segmenting the thrombus on brain MRI in patients with hyper-acute ischemic stroke. Proceedings of The 7nd International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 250:657-671 Available from https://proceedings.mlr.press/v250/ibarra24a.html.

Related Material