Semantic Mosaicing of Histo-Pathology Image Fragments using Visual Foundation Models

Stefan Brandstätter, Maximilan Köller, Philipp Seeböck, Alissa Blessing, Felicitas Oberndorfer, Svitlana Pochepnia, Helmut Prosch, Georg Langs
Proceedings of the MICCAI Workshop on Computational Pathology, PMLR 316:213-222, 2026.

Abstract

In histopathology, tissue samples are often larger than a standard microscope slide, making stitching of multiple fragments necessary to process entire structures such as tumors.Automated stitching is a prerequisite for scaling analysis, but is challenging due to possible tissue loss during preparation, inhomogeneous morphological distortion, staining inconsistencies, missing regions due to misalignment on the slide, or frayed tissue edges. This limits state-of-the-art stitching methods using boundary shape matching algorithms to reconstruct artificial whole mount slides (WMS). Here, we introduce SemanticStitcher using latent feature representations derived from a visual histopathology foundation model to identify neighboring areas in different fragments. Robust pose estimation based on a large number of semantic matching candidates derives a mosaic of multiple fragments to form the WMS. Experiments on three different histopathology datasets demonstrate that SemanticStitcher yields robust WMS mosaicing and consistently outperforms the state of the art in correct boundary matches.

Cite this Paper

BibTeX
@InProceedings{pmlr-v316-brandstatter26a, title = {Semantic Mosaicing of Histo-Pathology Image Fragments using Visual Foundation Models}, author = {Brandst\"{a}tter, Stefan and K\"{o}ller, Maximilan and Seeb\"{o}ck, Philipp and Blessing, Alissa and Oberndorfer, Felicitas and Pochepnia, Svitlana and Prosch, Helmut and Langs, Georg}, booktitle = {Proceedings of the MICCAI Workshop on Computational Pathology}, pages = {213--222}, year = {2026}, editor = {Studer, Linda and Ciompi, Francesco and Khalili, Nadieh and Faryna, Khrystyna and Faryna, Khrystyna and Yeong, Joe and Lau, Mai Chan and Chen, Hao and Liu, Ziyi and Brattoli, Biagio}, volume = {316}, series = {Proceedings of Machine Learning Research}, month = {27 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v316/main/assets/brandstatter26a/brandstatter26a.pdf}, url = {https://proceedings.mlr.press/v316/brandstatter26a.html}, abstract = {In histopathology, tissue samples are often larger than a standard microscope slide, making stitching of multiple fragments necessary to process entire structures such as tumors.Automated stitching is a prerequisite for scaling analysis, but is challenging due to possible tissue loss during preparation, inhomogeneous morphological distortion, staining inconsistencies, missing regions due to misalignment on the slide, or frayed tissue edges. This limits state-of-the-art stitching methods using boundary shape matching algorithms to reconstruct artificial whole mount slides (WMS). Here, we introduce SemanticStitcher using latent feature representations derived from a visual histopathology foundation model to identify neighboring areas in different fragments. Robust pose estimation based on a large number of semantic matching candidates derives a mosaic of multiple fragments to form the WMS. Experiments on three different histopathology datasets demonstrate that SemanticStitcher yields robust WMS mosaicing and consistently outperforms the state of the art in correct boundary matches.} }
Endnote
%0 Conference Paper %T Semantic Mosaicing of Histo-Pathology Image Fragments using Visual Foundation Models %A Stefan Brandstätter %A Maximilan Köller %A Philipp Seeböck %A Alissa Blessing %A Felicitas Oberndorfer %A Svitlana Pochepnia %A Helmut Prosch %A Georg Langs %B Proceedings of the MICCAI Workshop on Computational Pathology %C Proceedings of Machine Learning Research %D 2026 %E Linda Studer %E Francesco Ciompi %E Nadieh Khalili %E Khrystyna Faryna %E Khrystyna Faryna %E Joe Yeong %E Mai Chan Lau %E Hao Chen %E Ziyi Liu %E Biagio Brattoli %F pmlr-v316-brandstatter26a %I PMLR %P 213--222 %U https://proceedings.mlr.press/v316/brandstatter26a.html %V 316 %X In histopathology, tissue samples are often larger than a standard microscope slide, making stitching of multiple fragments necessary to process entire structures such as tumors.Automated stitching is a prerequisite for scaling analysis, but is challenging due to possible tissue loss during preparation, inhomogeneous morphological distortion, staining inconsistencies, missing regions due to misalignment on the slide, or frayed tissue edges. This limits state-of-the-art stitching methods using boundary shape matching algorithms to reconstruct artificial whole mount slides (WMS). Here, we introduce SemanticStitcher using latent feature representations derived from a visual histopathology foundation model to identify neighboring areas in different fragments. Robust pose estimation based on a large number of semantic matching candidates derives a mosaic of multiple fragments to form the WMS. Experiments on three different histopathology datasets demonstrate that SemanticStitcher yields robust WMS mosaicing and consistently outperforms the state of the art in correct boundary matches.
APA
Brandstätter, S., Köller, M., Seeböck, P., Blessing, A., Oberndorfer, F., Pochepnia, S., Prosch, H. & Langs, G.. (2026). Semantic Mosaicing of Histo-Pathology Image Fragments using Visual Foundation Models. Proceedings of the MICCAI Workshop on Computational Pathology, in Proceedings of Machine Learning Research 316:213-222 Available from https://proceedings.mlr.press/v316/brandstatter26a.html.

Related Material