Every year, around 500,000 people in Germany are diagnosed with cancer. Pathologists have a central responsibility in diagnostics: they analyze almost 40 million tissue samples every year, usually on the basis of a few histological sections selected on the basis of experience.
However, this traditional process has its limits - relevant structures remain hidden in the often extensive FFPE (formalin-fixed, paraffin-embedded) tissue blocks. This is precisely where the BMBF-funded research project HORUS (13GW0571D) comes in.
In our new project video, we provide insights into:
The perspectives of the researchers involved
The innovative imaging and analysis workflow
The future of AI-assisted cancer diagnostics and virtual histology
Virtual histology with micro-CT: How HORUS is rethinking cancer diagnostics
What does "virtual histology with micro-CT" mean?
The aim of virtual histology is to examine tissue samples three-dimensionally and non-destructively before they are cut microscopically. At the heart of this is micro-CT imaging - a high-resolution X-ray technique that uses phase contrast and absorption CT to visualize structures ranging from 10 µm fine capillaries to millimetre-sized germinal centers.
In combination with artificial intelligence (AI) and annotation tools, critical areas such as tumor margins, blood vessels or germinal centers can be precisely marked and analyzed. Pathology laboratories can thus specifically identify the relevant part of a tissue block before it is further processed under the microtome.
Why is this a breakthrough for pathology?
Significantly higher precision:
Thanks to the combination of virtual histology and micro-CT, the sampling error can be reduced by a factor of 1,000 to 5,000.
While current diagnoses are often based on just a few 2-micrometer sections, HORUS makes it possible to analyze the entire tumor volume - from centimeter-sized tumors to fine 10-25 µm structures such as capillaries or glands.Gentle diagnostics:
The FFPE tissue block is completely preserved so that it can continue to be used for molecular genetic analyses. Pathologists no longer have to decide which part of the sample should be sacrificed.More efficient workflows:
The 3D preview and AI-supported annotation allow relevant areas to be specifically identified, which will optimize working time in the future without additional effort - a decisive advantage in view of the shortage of specialist pathologists in Germany.Digital networking:
With the help of the software used, such as the ImFusion Suite, micro-CT data and conventional histology can be merged in a digital workflow that can be integrated into existing laboratory processes in the future.
Digital infrastructure: Our HORUS contribution
For virtual histology to work not only technologically, but also in everyday clinical practice, smooth processes are needed in the background.
In the HORUS project, we have incorporated our tried-and-tested workflow engine - which can be used to reliably map medical processes from digital order entry and AI-supported context data to structured reporting of findings.
A special feature: the graphic marking of the sampling site by the clinic user - automatically coded and clearly visible to pathologists.
The HORUS project partners
The HORUS project (funding reference 13GW0571D), funded by the German Federal Ministry of Education and Research (BMBF), brings together research, clinical practice and industry. The partners are pooling their expertise to bring virtual histology with micro-CT into practice.
Federal Ministry of Education and Research (BMBF)
Funds the project within the framework of medical technology to support innovative diagnostic procedures and enable their transfer to clinical application.
University Hospital Düsseldorf - Institute of Pathology
Provides clinical expertise and patient samples and tests the developed methods directly in cancer diagnostics, especially in the field of tumor characterization and tissue analysis.
Freiburg University Medical Center - Institute of Pathology
Works on the integration of micro-CT data into the pathological workflow and ensures that the methods meet clinical requirements.
ImFusion GmbH
Develops AI-supported software solutions for image fusion, segmentation and annotation so that pathologists can combine micro-CT and histology in a digital workflow.
Fraunhofer Institute for High-Speed Dynamics, EMI
Advances the technical development of micro-CT imaging and optimizes the preparation of tissue samples for phase-contrast-based imaging.
DORNER Health IT Solutions
provides support with technological expertise and data management so that the developed processes can be implemented in a scalable and practical manner.
Conclusion
HORUS is a pioneering approach to cancer diagnostics:
Virtual histology with micro-CT and artificial intelligence make it possible to see more, destroy less and diagnose faster - with maximum precision.