HistokatFusion combines the information from multiple images in histopathology using image registration. The technology has been used in biomarker research, AI development and to combine and evaluate different stains.
The algorithm behind HistokatFusion won the first place at the ANHIR challenge on histological image registation.
Automatic initial alignment of slide images. Most bright-field images do not require manual interactions.
Wouter Bulten et al., Epithelium segmentation using deep learning in H&E-stained prostate specimens with immunohistochemistry as reference standard. Nature Scientific Reports, 2019
Maschenka CA Balkenhol et al., Optimized tumour infiltrating lymphocyte assessment for triple negative breast cancer prognostics, The Breast, 2021
J. Lotz, N. Weiss, J. vd Laak, S. Heldmann, High-resolution Image Registration of Consecutive and Re-stained Sections in Histopathology. submitted, arxiv preprint, 2021
J. Borovec et al., ANHIR: Automatic Non-Rigid Histological Image Registration Challenge, IEEE Transactions on Medical Imaging, 2020
Integrate HistokatFusion from Python, C++ or use our serverless compute infrastructure on-demand.
HistokatFusion has been validated on thousands of differently stained image pairs.