Joining forces for pathology diagnostics with AI assistance: The EMPAIA initiative.

Over the past decade, artificial intelligence (AI) methods in pathology have advanced substantially. However, integration into routine clinical practice has been slow due to numerous challenges, including technical and regulatory hurdles in translating research results into clinical diagnostic products and the lack of standardized interfaces. The open and vendor-neutral EMPAIA initiative addresses these challenges. Here, we provide an overview of EMPAIA's achievements and lessons learned. EMPAIA integrates various stakeholders of the pathology AI ecosystem, i.e., pathologists, computer scientists, and industry. In close collaboration, we developed technical interoperability standards, recommendations for AI testing and product development, and explainability methods. We implemented the modular and open-source EMPAIA Platform and successfully integrated 14 AI-based image analysis apps from eight different vendors, demonstrating how different apps can use a single standardized interface. We prioritized requirements and evaluated the use of AI in real clinical settings with 14 different pathology laboratories in Europe and Asia. In addition to technical developments, we created a forum for all stakeholders to share information and experiences on digital pathology and AI. Commercial, clinical, and academic stakeholders can now adopt EMPAIA's common open-source interfaces, providing a unique opportunity for large-scale standardization and streamlining of processes. Further efforts are needed to effectively and broadly establish AI assistance in routine laboratory use. To this end, a sustainable infrastructure, the non-profit association EMPAIA International, has been established to continue standardization and support broad implementation and advocacy for an AI-assisted digital pathology future.

Explainability and causability in digital pathology.

The current move towards digital pathology enables pathologists to use artificial intelligence (AI)-based computer programmes for the advanced analysis of whole slide images. However, currently, the best-performing AI algorithms for image analysis are deemed black boxes since it remains - even to their developers - often unclear why the algorithm delivered a particular result. Especially in medicine, a better understanding of algorithmic decisions is essential to avoid mistakes and adverse effects on patients. This review article aims to provide medical experts with insights on the issue of explainability in digital pathology. A short introduction to the relevant underlying core concepts of machine learning shall nurture the reader's understanding of why explainability is a specific issue in this field. Addressing this issue of explainability, the rapidly evolving research field of explainable AI (XAI) has developed many techniques and methods to make black-box machine-learning systems more transparent. These XAI methods are a first step towards making black-box AI systems understandable by humans. However, we argue that an explanation interface must complement these explainable models to make their results useful to human stakeholders and achieve a high level of causability, i.e. a high level of causal understanding by the user. This is especially relevant in the medical field since explainability and causability play a crucial role also for compliance with regulatory requirements. We conclude by promoting the need for novel user interfaces for AI applications in pathology, which enable contextual understanding and allow the medical expert to ask interactive 'what-if'-questions. In pathology, such user interfaces will not only be important to achieve a high level of causability. They will also be crucial for keeping the human-in-the-loop and bringing medical experts' experience and conceptual knowledge to AI processes.

Recommendations on compiling test datasets for evaluating artificial intelligence solutions in pathology.

Artificial intelligence (AI) solutions that automatically extract information from digital histology images have shown great promise for improving pathological diagnosis. Prior to routine use, it is important to evaluate their predictive performance and obtain regulatory approval. This assessment requires appropriate test datasets. However, compiling such datasets is challenging and specific recommendations are missing. A committee of various stakeholders, including commercial AI developers, pathologists, and researchers, discussed key aspects and conducted extensive literature reviews on test datasets in pathology. Here, we summarize the results and derive general recommendations on compiling test datasets. We address several questions: Which and how many images are needed? How to deal with low-prevalence subsets? How can potential bias be detected? How should datasets be reported? What are the regulatory requirements in different countries? The recommendations are intended to help AI developers demonstrate the utility of their products and to help pathologists and regulatory agencies verify reported performance measures. Further research is needed to formulate criteria for sufficiently representative test datasets so that AI solutions can operate with less user intervention and better support diagnostic workflows in the future.

EMPAIA App Interface: An open and vendor-neutral interface for AI applications in pathology.

BACKGROUND AND OBJECTIVE: Artificial intelligence (AI) apps hold great potential to make pathological diagnoses more accurate and time efficient. Widespread use of AI in pathology is hampered by interface incompatibilities between pathology software. We studied the existing interfaces in order to develop the EMPAIA App Interface, an open standard for the integration of pathology AI apps. METHODS: The EMPAIA App Interface relies on widely-used web communication protocols and containerization. It consists of three parts: A standardized format to describe the semantics of an app, a mechanism to deploy and execute apps in computing environments, and a web API through which apps can exchange data with a host application. RESULTS: Five commercial AI app manufacturers successfully adapted their products to the EMPAIA App Interface and helped improve it with their feedback. Open source tools facilitate the adoption of the interface by providing reusable data access and scheduling functionality and enabling automatic validation of app compliance. CONCLUSIONS: Existing AI apps and pathology software can be adapted to the EMPAIA App Interface with little effort. It is a viable alternative to the proprietary interfaces of current software. If enough vendors join in, the EMPAIA App Interface can help to advance the use of AI in pathology.

Toward a methodical framework for comprehensively assessing forest multifunctionality.

Biodiversity-ecosystem functioning (BEF) research has extended its scope from communities that are short-lived or reshape their structure annually to structurally complex forest ecosystems. The establishment of tree diversity experiments poses specific methodological challenges for assessing the multiple functions provided by forest ecosystems. In particular, methodological inconsistencies and nonstandardized protocols impede the analysis of multifunctionality within, and comparability across the increasing number of tree diversity experiments. By providing an overview on key methods currently applied in one of the largest forest biodiversity experiments, we show how methods differing in scale and simplicity can be combined to retrieve consistent data allowing novel insights into forest ecosystem functioning. Furthermore, we discuss and develop recommendations for the integration and transferability of diverse methodical approaches to present and future forest biodiversity experiments. We identified four principles that should guide basic decisions concerning method selection for tree diversity experiments and forest BEF research: (1) method selection should be directed toward maximizing data density to increase the number of measured variables in each plot. (2) Methods should cover all relevant scales of the experiment to consider scale dependencies of biodiversity effects. (3) The same variable should be evaluated with the same method across space and time for adequate larger-scale and longer-time data analysis and to reduce errors due to changing measurement protocols. (4) Standardized, practical and rapid methods for assessing biodiversity and ecosystem functions should be promoted to increase comparability among forest BEF experiments. We demonstrate that currently available methods provide us with a sophisticated toolbox to improve a synergistic understanding of forest multifunctionality. However, these methods require further adjustment to the specific requirements of structurally complex and long-lived forest ecosystems. By applying methods connecting relevant scales, trophic levels, and above- and belowground ecosystem compartments, knowledge gain from large tree diversity experiments can be optimized.

Kinetic energy of Throughfall in subtropical forests of SE China - effects of tree canopy structure, functional traits, and biodiversity.

Throughfall kinetic energy (TKE) plays an important role in soil erosion in forests. We studied TKE as a function of biodiversity, functional diversity as well as structural stand variables in a secondary subtropical broad-leaved forest in the Gutianshan National Nature Reserve (GNNR) in south-east China, a biodiversity hotspot in the northern hemisphere with more than 250 woody species present. Using a mixed model approach we could identify significant effects of all these variables on TKE: TKE increased with rarefied tree species richness and decreased with increasing proportion of needle-leaved species and increasing leaf area index (LAI). Furthermore, for average rainfall amounts TKE was decreasing with tree canopy height whereas for high rainfall amounts this was not the case. The spatial pattern of throughfall was stable across several rain events. The temporal variation of TKE decreased with rainfall intensity and increased with tree diversity. Our results show that more diverse forest stands over the season have to cope with higher cumulative raindrop energy than less diverse stands. However, the kinetic energy (KE) of one single raindrop is less predictable in diverse stands since the variability in KE is higher. This paper is the first to contribute to the understanding of the ecosystem function of soil erosion prevention in diverse subtropical forests.