Orientation Normalization of Multi-Stain Skin Tissue Cross-Sections

Ema Topolnjak, Evi Paulides, Willeke A. M. Blokx, Mitko Veta, Ruben T. Lucassen
Proceedings of The 9th International Conference on Medical Imaging with Deep Learning, PMLR 315:322-341, 2026.

Abstract

Efficient examination of skin tissue specimens is key for pathologists to keep up with an increasing workload. Normalizing the orientation of tissue cross-sections before manual assessment could contribute to a more streamlined digital workflow. In this study, we compare multiple deep learning-based approaches for predicting the rotation angle required to correct the misorientation of skin tissue cross-sections. The models were developed and evaluated using a dataset of 10,649 H&E-stained and 9,731 IHC-stained cross-section images from specimens with melanocytic lesions. Our results show that framing rotation angle prediction as a classification task with the circular target space divided into separate classes performed best, reaching mean absolute errors of 2.77$^\circ$ and 3.56$^\circ$ on the test sets of H&E and IHC-stained cross-sections, respectively, approaching the level of human annotators. Automated orientation normalization, when implemented in whole slide image viewers, could make tissue examination more efficient and convenient for pathologists, while also serving as a valuable preprocessing step for the development of position-aware or multi-stain deep learning models.

Cite this Paper

BibTeX
@InProceedings{pmlr-v315-topolnjak26a, title = {Orientation Normalization of Multi-Stain Skin Tissue Cross-Sections}, author = {Topolnjak, Ema and Paulides, Evi and Blokx, Willeke A. M. and Veta, Mitko and Lucassen, Ruben T.}, booktitle = {Proceedings of The 9th International Conference on Medical Imaging with Deep Learning}, pages = {322--341}, year = {2026}, editor = {Huo, Yuankai and Gao, Mingchen and Kuo, Chang-Fu and Jin, Yueming and Deng, Ruining}, volume = {315}, series = {Proceedings of Machine Learning Research}, month = {08--10 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v315/main/assets/topolnjak26a/topolnjak26a.pdf}, url = {https://proceedings.mlr.press/v315/topolnjak26a.html}, abstract = {Efficient examination of skin tissue specimens is key for pathologists to keep up with an increasing workload. Normalizing the orientation of tissue cross-sections before manual assessment could contribute to a more streamlined digital workflow. In this study, we compare multiple deep learning-based approaches for predicting the rotation angle required to correct the misorientation of skin tissue cross-sections. The models were developed and evaluated using a dataset of 10,649 H&E-stained and 9,731 IHC-stained cross-section images from specimens with melanocytic lesions. Our results show that framing rotation angle prediction as a classification task with the circular target space divided into separate classes performed best, reaching mean absolute errors of 2.77$^\circ$ and 3.56$^\circ$ on the test sets of H&E and IHC-stained cross-sections, respectively, approaching the level of human annotators. Automated orientation normalization, when implemented in whole slide image viewers, could make tissue examination more efficient and convenient for pathologists, while also serving as a valuable preprocessing step for the development of position-aware or multi-stain deep learning models.} }
Endnote
%0 Conference Paper %T Orientation Normalization of Multi-Stain Skin Tissue Cross-Sections %A Ema Topolnjak %A Evi Paulides %A Willeke A. M. Blokx %A Mitko Veta %A Ruben T. Lucassen %B Proceedings of The 9th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2026 %E Yuankai Huo %E Mingchen Gao %E Chang-Fu Kuo %E Yueming Jin %E Ruining Deng %F pmlr-v315-topolnjak26a %I PMLR %P 322--341 %U https://proceedings.mlr.press/v315/topolnjak26a.html %V 315 %X Efficient examination of skin tissue specimens is key for pathologists to keep up with an increasing workload. Normalizing the orientation of tissue cross-sections before manual assessment could contribute to a more streamlined digital workflow. In this study, we compare multiple deep learning-based approaches for predicting the rotation angle required to correct the misorientation of skin tissue cross-sections. The models were developed and evaluated using a dataset of 10,649 H&E-stained and 9,731 IHC-stained cross-section images from specimens with melanocytic lesions. Our results show that framing rotation angle prediction as a classification task with the circular target space divided into separate classes performed best, reaching mean absolute errors of 2.77$^\circ$ and 3.56$^\circ$ on the test sets of H&E and IHC-stained cross-sections, respectively, approaching the level of human annotators. Automated orientation normalization, when implemented in whole slide image viewers, could make tissue examination more efficient and convenient for pathologists, while also serving as a valuable preprocessing step for the development of position-aware or multi-stain deep learning models.
APA
Topolnjak, E., Paulides, E., Blokx, W.A.M., Veta, M. & Lucassen, R.T.. (2026). Orientation Normalization of Multi-Stain Skin Tissue Cross-Sections. Proceedings of The 9th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 315:322-341 Available from https://proceedings.mlr.press/v315/topolnjak26a.html.

Related Material