Committees
The EMPAIA Committees (working groups) are an important part of the ecosystem. They are open to all stakeholders in the ecosystem. The results of the work will be made available in the form of a white paper.
The committees and their goals are briefly presented below. The committee work is mainly done electronically. Face-to-face meetings take place in the context of national and international conferences
How can I sign up to participate in one or more committees?
Please send an email with your contact details to Nora Charlotte Pohlan: info@empaia.org
Standardization
Aims of the Committee
The standardization of interfaces and data formats provides a substantial basis for the interoperability of systems in digital pathology in general and for the use of AI applications in routine diagnosis in particular. On an international level, standardization is mainly driven by IHE PaLM and DICOM WG 26. The specific issues arising from the use of AI solutions have so far only been marginally addressed. The EMPAIA standardization committee, in close coordination with IHE and DICOM, wants to make standard proposals for the relevant interfaces in the context of the use of AI solutions and support their implementation. The goal is to provide interfaces and reference implementations quickly. These are to be established with the goal of international standardization in coordination with IHE and DICOM. If standards are missing, preliminary (coordinated, standardizable) solutions will be implemented in the reference centers to make practical use of AI solutions and to gain experience. Results will be published in the form of a white paper.
Topics
-
Technical interfaces
-
Data formats
-
IHE (Workflows)
-
DICOM (image management)
-
Integration of algorithms into the workflow (workstations)
-
Interoperability of algorithms
Chairs:
Gunter Haroske, Bundesverband Deutscher Pathologen e. V. (BDP)
Norman Zerbe, Institut für Digitale Pathologie, Charité - Universitätsmedizin Berlin
Profitability & Accounting
Aims of the Committee
Experts agree that AI solutions will strongly influence and change pathology. Digitization will lay the decisive foundation for the use of AI methods in pathology. This will provide decisive impulses for quality assurance in pathology, further education, but also for interdisciplinary cooperation with clinicians. AI methods affect morphological and molecular pathological diagnostics equally and stimulate their combination. In order to advance these developments, questions of economic efficiency, billability of services and the creation of the necessary conditions must be solved. To this end, the billability committee works out the technical basis for upcoming "political" solutions in cooperation and with the participation of the Professional Association of German Pathologists, representatives of health insurance companies, chambers, physicians' associations and with the involvement of experts. In addition to the groups mentioned above, representatives from the EMPAIA reference centres and industrial partners are also involved.
Topics
-
Accounting principles
-
Changes through digitisation and AI in the world of work
-
Economic consequences of technical changes
-
Lobbying for pathological service providers
Chairs:
Jörg Kriegsmann, MVZ für Histologie, Zytologie und Molekulare Diagnostik
Wolfgang Schmitt, Institut für Pathologie, Charité - Universitätsmedizin Berlin
Certification of algorithms and systems
Aims of the Committee
AI algorithms must be certified (IVDR). The independent validation of AI algorithms and solutions is a necessary precondition for this. In the committee "certification of algorithms and systems", together with all stakeholders in pathology, it is to be examined how the robustness, integrity and last but not least the explainability of algorithms and systems can be realized. In addition to the AI algorithms in the foreground, systems such as specimen scanners and other diagnosis-related systems (e.g. tissue processors) are at the centre of the work. As a result, coordinated inspection concepts as well as reference data sets and implementations based on them are to be realized, which support manufacturers in preparing their products for certification. Furthermore, questions of certification and accreditation of facilities will be considered, as laboratory processes can have an impact on the application of AI algorithms.
Topics
-
Validation
-
Certification
-
Accreditation of centers in the context of digital pathology use
-
Certification as a medical product
-
Robustness and Integrity
Chairs:
André Homeyer, Fraunhofer Institut für Digitale Medizin MEVIS
Christian Geissler, TU Berlin, DAI Labor
Legal issues for the use of digital methods and AI in pathology
Aims of the Committee
Who is liable? - The introduction of AI algorithms into clinical diagnostics also requires the examination and, if necessary, adaptation of the legal framework. The aim is to establish legal certainty in the use of these new methods. This includes identifying and naming deficits and triggering solutions. An important goal of the committee is therefore, in addition to fundamental aspects such as data protection and data security within the framework of the EMPAIA platform, also to address the question of liability and to evaluate with national and international experts the effects of the use of AI solutions on the German, but also the pan-European market. Furthermore, this committee also provides the framework for addressing cross-cutting legal issues from the other committees.
Topics
-
Data protection & data security
-
Liability
-
Effects of a pan-European market
-
Cross-cutting issues from all other bodies
Chairs:
Peter Schüller, Biotronik SE & Co. KG
Friederike Gräfin von Brühl, K&L Gates
Studies
Aims of the Committee
Whether it is the development of new solutions, the validation of existing solutions, the analysis of technical influences on the quality of solutions - studies play a decisive role everywhere. With the EMPAIA reference centres, EMPAIA has the possibility to organise and conduct multicentre studies, which are to be used in a targeted manner to solve fundamental questions (e.g. building up validation data sets, determining and minimising the influence of technical boundary conditions, e.g. laboratory and scanner). For this purpose, there are excellent conditions for the involvement of companies in digital pathology and AI. Due to digitization of pathology and the introduction of AI in pathology a transformation of the field is taking place. The need for new diagnostic parameters is very high. A current example is the marker PD-L1, which plays an increasingly important role in precision oncology. Solutions for standardising the interpretation of PD-L1 immunostains are urgently needed. Within the EMPAIA project there will be studies on this, as well as on other immunostains, which are prerequisites for targeted therapies and have already been verified in ring trials, such as estrogen receptor (ER), progesterone receptor (PR), HER2 and CD19. The EMPAIA project also offers excellent opportunities for joint scientific studies due to the consortium's versatile partners in diagnostic pathology, industry and professional associations.
Topics
-
Multicentre studies involving the reference centres
-
Structure of validation data sets
-
Initiation of sponsored studies
-
Participation in national or international studies
-
Establishment of 'publicly available data sets' for tasks of general interest
-
Structuring of data records
-
Studies on the quantification of new histological markers
Chairs:
Korinna Jöhrens, Uniklinikum Dresden
Peter Boor, Universitätsklinikum RWTH Aachen