MLV2-Net: Rater-Based Majority-Label Voting for Consistent Meningeal Lymphatic Vessel Segmentation

Fabian Bongratz, Markus Karmann, Adrian Holz, Moritz Bonhoeffer, Viktor Neumaier, Sarah Deli, Benita Schmitz-Koep, Claus Zimmer, Christian Sorg, Melissa Thalhammer, Dennis M Hedderich, Christian Wachinger
Proceedings of the 4th Machine Learning for Health Symposium, PMLR 259:143-153, 2025.

Abstract

Meningeal lymphatic vessels (MLVs) are responsible for the drainage of waste products from the human brain. An impairment in their functionality has been associated with aging as well as brain disorders like multiple sclerosis and Alzheimer’s disease. However, MLVs have only recently been described for the first time in magnetic resonance imaging (MRI), and their ramified structure renders manual segmentation particularly difficult. Further, as there is no consistent notion of their appearance, human-annotated MLV structures contain a high inter-rater variability that most automatic segmentation methods cannot take into account. In this work, we propose a new rater-aware training scheme for the popular nnUNet model, and we explore rater-based ensembling strategies for accurate and consistent segmentation of MLVs. This enables us to boost nnU-Net’s performance while obtaining explicit predictions in different annotation styles and a rater-based uncertainty estimation. Our final model, MLV2-Net, achieves a Dice similarity coefficient of 0.806 with respect to the human reference standard. The model further matches the human inter-rater reliability and replicates age-related associations with MLV volume.

Cite this Paper


BibTeX
@InProceedings{pmlr-v259-bongratz25a, title = {MLV2-Net: Rater-Based Majority-Label Voting for Consistent Meningeal Lymphatic Vessel Segmentation}, author = {Bongratz, Fabian and Karmann, Markus and Holz, Adrian and Bonhoeffer, Moritz and Neumaier, Viktor and Deli, Sarah and Schmitz-Koep, Benita and Zimmer, Claus and Sorg, Christian and Thalhammer, Melissa and Hedderich, Dennis M and Wachinger, Christian}, booktitle = {Proceedings of the 4th Machine Learning for Health Symposium}, pages = {143--153}, year = {2025}, editor = {Hegselmann, Stefan and Zhou, Helen and Healey, Elizabeth and Chang, Trenton and Ellington, Caleb and Mhasawade, Vishwali and Tonekaboni, Sana and Argaw, Peniel and Zhang, Haoran}, volume = {259}, series = {Proceedings of Machine Learning Research}, month = {15--16 Dec}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v259/main/assets/bongratz25a/bongratz25a.pdf}, url = {https://proceedings.mlr.press/v259/bongratz25a.html}, abstract = {Meningeal lymphatic vessels (MLVs) are responsible for the drainage of waste products from the human brain. An impairment in their functionality has been associated with aging as well as brain disorders like multiple sclerosis and Alzheimer’s disease. However, MLVs have only recently been described for the first time in magnetic resonance imaging (MRI), and their ramified structure renders manual segmentation particularly difficult. Further, as there is no consistent notion of their appearance, human-annotated MLV structures contain a high inter-rater variability that most automatic segmentation methods cannot take into account. In this work, we propose a new rater-aware training scheme for the popular nnUNet model, and we explore rater-based ensembling strategies for accurate and consistent segmentation of MLVs. This enables us to boost nnU-Net’s performance while obtaining explicit predictions in different annotation styles and a rater-based uncertainty estimation. Our final model, MLV2-Net, achieves a Dice similarity coefficient of 0.806 with respect to the human reference standard. The model further matches the human inter-rater reliability and replicates age-related associations with MLV volume.} }
Endnote
%0 Conference Paper %T MLV2-Net: Rater-Based Majority-Label Voting for Consistent Meningeal Lymphatic Vessel Segmentation %A Fabian Bongratz %A Markus Karmann %A Adrian Holz %A Moritz Bonhoeffer %A Viktor Neumaier %A Sarah Deli %A Benita Schmitz-Koep %A Claus Zimmer %A Christian Sorg %A Melissa Thalhammer %A Dennis M Hedderich %A Christian Wachinger %B Proceedings of the 4th Machine Learning for Health Symposium %C Proceedings of Machine Learning Research %D 2025 %E Stefan Hegselmann %E Helen Zhou %E Elizabeth Healey %E Trenton Chang %E Caleb Ellington %E Vishwali Mhasawade %E Sana Tonekaboni %E Peniel Argaw %E Haoran Zhang %F pmlr-v259-bongratz25a %I PMLR %P 143--153 %U https://proceedings.mlr.press/v259/bongratz25a.html %V 259 %X Meningeal lymphatic vessels (MLVs) are responsible for the drainage of waste products from the human brain. An impairment in their functionality has been associated with aging as well as brain disorders like multiple sclerosis and Alzheimer’s disease. However, MLVs have only recently been described for the first time in magnetic resonance imaging (MRI), and their ramified structure renders manual segmentation particularly difficult. Further, as there is no consistent notion of their appearance, human-annotated MLV structures contain a high inter-rater variability that most automatic segmentation methods cannot take into account. In this work, we propose a new rater-aware training scheme for the popular nnUNet model, and we explore rater-based ensembling strategies for accurate and consistent segmentation of MLVs. This enables us to boost nnU-Net’s performance while obtaining explicit predictions in different annotation styles and a rater-based uncertainty estimation. Our final model, MLV2-Net, achieves a Dice similarity coefficient of 0.806 with respect to the human reference standard. The model further matches the human inter-rater reliability and replicates age-related associations with MLV volume.
APA
Bongratz, F., Karmann, M., Holz, A., Bonhoeffer, M., Neumaier, V., Deli, S., Schmitz-Koep, B., Zimmer, C., Sorg, C., Thalhammer, M., Hedderich, D.M. & Wachinger, C.. (2025). MLV2-Net: Rater-Based Majority-Label Voting for Consistent Meningeal Lymphatic Vessel Segmentation. Proceedings of the 4th Machine Learning for Health Symposium, in Proceedings of Machine Learning Research 259:143-153 Available from https://proceedings.mlr.press/v259/bongratz25a.html.

Related Material