Two Heads Are Enough: DualU-Net, a Fast and Efficient Architecture for Nuclei Instance Segmentation

David Anglada-Rotger, Berta Jansat, Ferran Marques, Montse Pardàs
Proceedings of The 8th International Conference on Medical Imaging with Deep Learning, PMLR 301:15-29, 2026.

Abstract

Accurate detection and classification of cell nuclei in histopathological images are critical for both clinical diagnostics and large-scale digital pathology workflows. In this work, we introduce DualU-Net, a fully convolutional, multi-task architecture designed to streamline nuclei classification and segmentation. Unlike the widely adopted three-decoder paradigm of HoVer-Net, DualU-Net employs only two output heads: a segmentation decoder that predicts pixel-wise classification maps and a detection decoder that estimates Gaussian-based centroid density maps. By leveraging these two outputs, our model effectively reconstructs instance-level segmentations. The proposed architecture results in significantly faster inference, reducing processing time by up to x5 compared to HoVer-Net, while achieving classification and detection performance comparable to State-of-the-Art models. Additionally, our approach demonstrates greater computational efficiency than CellViT and NuLite. We further show that DualU-Net is more robust to staining variations, a common challenge in digital pathology workflows. The model has been successfully deployed in clinical settings as part of the DigiPatICS initiative, operating across eight hospitals within the Institut Catal{à} de la Salut (ICS) network, highlighting the practical viability of DualU-Net as an efficient and scalable solution for nuclei segmentation and classification in real-world pathology applications. The code and pretrained model weights are publicly available on https://github.com/davidanglada/DualU-Net.

Cite this Paper

BibTeX
@InProceedings{pmlr-v301-anglada-rotger26a, title = {Two Heads Are Enough: DualU-Net, a Fast and Efficient Architecture for Nuclei Instance Segmentation}, author = {Anglada-Rotger, David and Jansat, Berta and Marques, Ferran and Pard\`as, Montse}, booktitle = {Proceedings of The 8th International Conference on Medical Imaging with Deep Learning}, pages = {15--29}, year = {2026}, editor = {Tasdizen, Tolga and Elhabian, Shireen and Summers, Ronald and Chen, Chen and Koch, Lisa and Zhuang, Yan}, volume = {301}, series = {Proceedings of Machine Learning Research}, month = {09--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v301/main/assets/anglada-rotger26a/anglada-rotger26a.pdf}, url = {https://proceedings.mlr.press/v301/anglada-rotger26a.html}, abstract = {Accurate detection and classification of cell nuclei in histopathological images are critical for both clinical diagnostics and large-scale digital pathology workflows. In this work, we introduce DualU-Net, a fully convolutional, multi-task architecture designed to streamline nuclei classification and segmentation. Unlike the widely adopted three-decoder paradigm of HoVer-Net, DualU-Net employs only two output heads: a segmentation decoder that predicts pixel-wise classification maps and a detection decoder that estimates Gaussian-based centroid density maps. By leveraging these two outputs, our model effectively reconstructs instance-level segmentations. The proposed architecture results in significantly faster inference, reducing processing time by up to x5 compared to HoVer-Net, while achieving classification and detection performance comparable to State-of-the-Art models. Additionally, our approach demonstrates greater computational efficiency than CellViT and NuLite. We further show that DualU-Net is more robust to staining variations, a common challenge in digital pathology workflows. The model has been successfully deployed in clinical settings as part of the DigiPatICS initiative, operating across eight hospitals within the Institut Catal{à} de la Salut (ICS) network, highlighting the practical viability of DualU-Net as an efficient and scalable solution for nuclei segmentation and classification in real-world pathology applications. The code and pretrained model weights are publicly available on https://github.com/davidanglada/DualU-Net.} }
Endnote
%0 Conference Paper %T Two Heads Are Enough: DualU-Net, a Fast and Efficient Architecture for Nuclei Instance Segmentation %A David Anglada-Rotger %A Berta Jansat %A Ferran Marques %A Montse Pardàs %B Proceedings of The 8th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2026 %E Tolga Tasdizen %E Shireen Elhabian %E Ronald Summers %E Chen Chen %E Lisa Koch %E Yan Zhuang %F pmlr-v301-anglada-rotger26a %I PMLR %P 15--29 %U https://proceedings.mlr.press/v301/anglada-rotger26a.html %V 301 %X Accurate detection and classification of cell nuclei in histopathological images are critical for both clinical diagnostics and large-scale digital pathology workflows. In this work, we introduce DualU-Net, a fully convolutional, multi-task architecture designed to streamline nuclei classification and segmentation. Unlike the widely adopted three-decoder paradigm of HoVer-Net, DualU-Net employs only two output heads: a segmentation decoder that predicts pixel-wise classification maps and a detection decoder that estimates Gaussian-based centroid density maps. By leveraging these two outputs, our model effectively reconstructs instance-level segmentations. The proposed architecture results in significantly faster inference, reducing processing time by up to x5 compared to HoVer-Net, while achieving classification and detection performance comparable to State-of-the-Art models. Additionally, our approach demonstrates greater computational efficiency than CellViT and NuLite. We further show that DualU-Net is more robust to staining variations, a common challenge in digital pathology workflows. The model has been successfully deployed in clinical settings as part of the DigiPatICS initiative, operating across eight hospitals within the Institut Catal{à} de la Salut (ICS) network, highlighting the practical viability of DualU-Net as an efficient and scalable solution for nuclei segmentation and classification in real-world pathology applications. The code and pretrained model weights are publicly available on https://github.com/davidanglada/DualU-Net.
APA
Anglada-Rotger, D., Jansat, B., Marques, F. & Pardàs, M.. (2026). Two Heads Are Enough: DualU-Net, a Fast and Efficient Architecture for Nuclei Instance Segmentation. Proceedings of The 8th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 301:15-29 Available from https://proceedings.mlr.press/v301/anglada-rotger26a.html.

Related Material