Ebook: Perspectives on Digital Pathology

EURO-TELEPATH (Telepathology Network in Europe) is a European Cooperation in ScienceTechnology (COST) Actionlabeled IC0604. It is aimed at fostering collaboration between European research groups working on an adequate technological framework for the management of multimedia electronic healthcare records covering both data and images, but also on standards and applications for processing such multimedia information in anatomic pathology, and communicating it to the clinical domain. The standardization work is performed in close collaboration with related organizations dedicated to the development of international standards (SDOs).

This research project has been launched at a meeting in Brussels on 6 November 2007, and ran until 5 November 2011. Throughout those four years, EURO-TELE-PATH has been consolidating the most renowned research references in the field of informatics, as applied to anatomic pathology, in order to develop, with the support of national and other European programmes and projects, standards and solutions necessary to represent, interpret, browse and retrieve digital medical images while preserving their diagnostic quality for clinical purposes, education and research.

EURO-TELEPATH has been supporting the exchange of young researchers to establish new areas of scientific collaboration, and complementing multidisciplinary publications and research theses on running prototypes in hospitals, laboratories and institutions. These areas of research were related to digital pathology, telepathology, medical informatics, image analysis, business process modeling, and the evaluation of technology.

At the end of the project, we asked the most relevant researchers in the field of digital pathology – most of whom had been active members of EURO-TELEPATH – to contribute to a book that would cover their main research initiatives. Perspectives on Digital Pathology is a book that compiles the main research efforts of the European COST Action IC0604 consortium, and which was presented as the project's final publication to the COST office. As well as rounding up the presentation and being as comprehensive as possible, it also includes a selection of papers from American and Japanese researchers working in the same area of research.

This book is divided into six Parts.

In Part I, “Introduction”, the COST instruments supporting cooperation among scientists and researchers across Europe and the activities of COST Action IC0604 are introduced. Part II, “State of the Art in Pathology”, contains five chapters. In the first chapter, the state of the art and trends in digital pathology are described, followed in Chapter 2 by a presentation of the Telemedicine Programme in Alentejo, Portugal. Chapter 3 provides a discussion about paradigm changes in health towards personalized, ubiquitous health services in pathology, and Chapter 4 presents a system architectural approach to telepathology interoperability. Chapter 5 deals with the consequences of the digital for pathology professionals, and the use of digital slides for education concludes this section on the “State of the Art in Digital Pathology”.

Part III, “Pathology Business Modeling”, presents the state of the art in pathology business process analysis, modeling, design and optimization. It also describes in detail a proposal for advanced pathology business process models.

Part IV, “Standards and Specifications in Pathology”, offers a complete compilation of current image management, report management and terminology standards in pathology. Given the importance of ontologies for semantic interoperability, a chapter about SNOMED CT in pathology has been included.

Part V, “Images: Analysis, Processing, Retrieval and Management”, provides two contributions on digital pathology in personalized cancer therapy, including an example of an automated image analysis methodology and the effects of image compression. Some work on content based image retrieval (CBIR) follows.

Part VI, “Technology and Automation in Pathology”, describes an advanced disrupting scanning technology developed in a European country. The second chapter addresses the relevance and importance of security and privacy services in running any pathology solution. The last chapter in this section concerns grid architecture and components which play an important role in diagnostic pathology.

Part VII, “Strategic developments and emerging research”, comprises three relevant chapters that describe the future of pathology technology, including new user interfaces and fast viewing systems, three-dimensional (3D)-reconstruction in cancer pathology study, and a completely automated tissue sectioning system for tissue blocks.

The editors would like to thank all the contributing authors for their excellent work, which has made this final project achievement possible. Furthermore, they are indebted to the COST office for sponsoring the publication of this book.

Marcial García-Rojo, Bernd Blobel, Arvydas Laurinavicius

(Editors)

The COST Action IC0604 “Telepathology Network in Europe” (EURO-TELEPATH) is a European COST Action that has been running from 2007 to 2011. COST Actions are funded by the COST (European Cooperation in the field of Scientific and Technical Research) Agency, supported by the Seventh Framework Programme for Research and Technological Development (FP7), of the European Union. EURO-TELEPATH's main objectives were evaluating and validating the common technological framework and communication standards required to access, transmit and manage digital medical records by pathologists and other medical professionals in a networked environment. The project was organized in four working groups. orking Group 1 “Business modeling in pathology” has designed main pathology processes – Frozen Study, Formalin Fixed Specimen Study, Telepathology, Cytology, and Autopsy –using Business Process Modeling Notation (BPMN). orking Group 2 “Informatics standards in pathology” has been dedicated to promoting the development and application of informatics standards in pathology, collaborating with Integrating the Healthcare Enterprise (IHE), Digital Imaging and Communications in Medicine (DICOM), Health Level Seven (HL7), and other standardization bodies. Working Group 3 “Images: Analysis, Processing, Retrieval and Management” worked on the use of virtual or digital slides that are fostering the use of image processing and analysis in pathology not only for research purposes, but also in daily practice. Working Group 4 “Technology and Automation in Pathology” was focused on studying the adequacy of current existing technical solutions, including, e.g., the quality of images obtained by slide scanners, or the efficiency of image analysis applications. Major outcome of this action are the collaboration with international health informatics standardization bodies to foster the development of standards for digital pathology, offering a new approach for workflow analysis, based in business process modeling. Health terminology standardization research has become a topic of high interest. Future research work should focus on standardization of automatic image analysis and tissue microarrays imaging.

Anatomic pathology is a medical specialty where both information management systems and digital images systems paly a most important role. Digital pathology is a new concept that considers all uses of this information, including diagnosis, biomedical research and education. Virtual microscopy or whole slide imaging, resulting in digital slides, is an outreaching technology in anatomic pathology. Limiting factors in the expansion of virtual microscopy are formidable storage dimension, scanning speed, quality of image and cultural change. Anatomic pathology data and images should be an important part of the patient electronic health records as well as of clinical data warehouse, epidemiological or biomedical research databases, and platforms dedicated to translational medicine. Integrating anatomic pathology to the “healthcare enterprise” can only be achieved using existing and emerging medical informatics standards like Digital Imaging and Communications in Medicine (DICOM®¹), Health Level Seven (HL7®), and Systematized Nomenclature of Medicine-Clinical Terms (SNOMED CT®), following the recommendations of Integrating the Healthcare Enterprise (IHE®). The consequences of the full digitalization of pathology departments are hard to foresee, but short term issues have arisen that imply interesting challenges for health care standards bodies

Alentejo – one of five Portuguese continental regions – faces major problems impacting the health and social system of the region. Here, the low population density, the low educational and income level as well as an aging population have to be mentioned. Faced with the task of ensuring equal access to healthcare for all its inhabitants, the regional health authorities created the telemedicine program.

From 1998 until 2000, the program developed in an experimental fashion, with teleconsultations involving a number of providers: primary health care centers, regional hospitals, and central hospitals. Between 2000 and 2010, there were a total of 135,000 telemedicine acts including teleconsultations, teleradiology (computerised tomography and x-rays), ultrasound telemedicine and telepathology. Presently, the network comprises 20 health centers and 6 hospitals, covering 4 districts. The platform is composed of high resolution videoconferencing equipment, software with patients' clinical records, an image archive, and a number of peripherals, such as electronic dermatoscopes and phonendoscopes. Teleconsultations are provided by fifteen medical specialties, across 3 district hospitals, ranging from neurology to pediatric surgery.

In 2008, health authorities started the telelearning program, initially using point to point videoconferencing, and by the end of 2010, 848 healthcare professionals, across 52 locations, had participated in remote learning sessions, covering topics from chronic wound treatment, to infection control, to medical error. As of 2011, point to multipoint telelearning is also in operation.

This paper provides an overview of the telemedicine program in Alentejo, including both infrastructure and operations. Preliminary results of an ongoing evaluation of the impact of teleconsultations on key indicators of the regional healthcare system are also presented (including current utilization and plans for future expansion). This article builds on the experience acquired throughout a decade of using telemedicine on an everyday basis, in a context of remarkable challenges in the delivery of accessible, equitable and quality healthcare services.

For the sake of safety and quality of care as well as efficiency of care processes, health systems undergo a paradigm change towards personalized, ubiquitous, health services. This change includes preventive and predictive medicine based on advanced translational medicine. Here we introduce domain-specific, organizational, and technical paradigms, requirements and solutions for personalized, ubiquitous, care. Emphasizing the formal aspects of modeling and implementing Telehealth and personal health (pHealth) interoperability and the entailed multidisciplinary integration, and illustrate the drivers behind and benefits of personalized medicine with a specific focus on the changing trends and impact on pathology, especially emphasizing Telepathology.

eHealth and pHealth, instantiated, e.g., in telepathology solutions, have to meet advanced interoperability challenges. Technologically supported by pervasive computing and even autonomic computing, pHealth covers many domains, scientifically managed by specialized disciplines using their specific ontologies. Therefore, semantic interoperability has to advance from a communication protocol to knowledge coordination and sharing, deploying advanced system architectures. Based on long-term work of scientific institutions and SDOs dedicated to system architectures, an interoperability framework is presented, integrating existing, emerging and even future specifications and standards for comprehensive interoperability of health and social services. The methodology is proven in many health information systems implementation and standard developments projects.

New opportunities and the adoption of digital technologies will transform the way pathology professionals and services work. Many areas of our daily life as well as medical professions have experienced this change already which has resulted in a paradigm shift in many activities. Pathology is an image-based discipline, therefore, arrival of digital imaging into this domain promises major shift in our work and required mentality. Recognizing the physical and digital duality of the pathology workflow, we can prepare for the imminent increase of the digital component, synergize and enjoy its benefits. Development of a new generation of laboratory information systems along with seamless integration of digital imaging, decision-support, and knowledge databases will enable pathologists to work in a distributed environment. The paradigm of “cloud pathology” is proposed as an ultimate vision of digital pathology workstations plugged into the integrated multidisciplinary patient care systems.

The use of digital slides for educational purposes (both for medical students and during pathology traineeships) will eventually accelerate the acceptance of digital pathology in general. This chapter describes the advantages of using digital slides especially for education. Also the requirements for using digital slides for this purpose are evaluated, including software requirements, the slide scanner and the IT infrastructure needed to provide a robust infrastructure to end users.

The importance of the process point of view is not restricted to a specific enterprise sector. In the field of health, as a result of the nature of the service offered, health institutions' processes are also the basis for decision making which is focused on achieving their objective of providing quality medical assistance. In this chapter the application of business process modelling – using the Business Process Modelling Notation (BPMN) standard is described. Main challenges of business process modelling in healthcare are the definition of healthcare processes, the multi-disciplinary nature of healthcare, the flexibility and variability of the activities involved in health care processes, the need of interoperability between multiple information systems, and the continuous updating of scientific knowledge in healthcare.

For analyzing current workflows and processes, for improving them, for quality management and quality assurance, for integrating hardware and software components, but also for education, training and communication between different domains' experts, modeling business process in a pathology department is inevitable. The authors highlight three main processes in pathology: general diagnostic, cytology diagnostic, and autopsy. In this chapter, those processes are formally modeled and described in detail. Finally, specialized processes such as immunohistochemistry and frozen section have been considered.

For making medical decisions, healthcare professionals require that all necessary information is both correct and easily available. Collaborative Digital Anatomic Pathology refers to the use of information technology that supports the creation and sharing or exchange of information, including data and images, during the complex workflow performed in an Anatomic Pathology department from specimen reception to report transmission and exploitation. Collaborative Digital Anatomic Pathology is supported by standardization efforts toward knowledge representation for sharable and computable clinical information. The goal of the international integrating the Healthcare Enterprise (IHE) initiative is precisely specifying how medical informatics standards should be implemented to meet specific health care needs and making systems integration more efficient and less expensive. The IHE Anatomic Pathology initiative was launched to implement the best use of medical informatics standards in order to produce, share and exchange machine-readable structured reports and their evidences (including whole slide images) within hospitals and across healthcare facilities. DICOM supplements 122 and 145 provide flexible object information definitions dedicated respectively to specimen description and WSI acquisition, storage and display. The profiles “Anatomic Pathology Reporting for Public Health” (ARPH) and “Anatomic Pathology Structured Report” (APSR) provide standard templates and transactions for sharing or exchanging structured reports in which textual observations - encoded using PathLex, an international controlled vocabulary currently being mapped to SNOMED CT concepts - may be bound to digital images or regions of interest in images. Current implementations of IHE Anatomic Pathology profiles in North America, France and Spain demonstrate the applicability of recent advances in standards for Collaborative Digital Anatomic Pathology. The use of machine-readable format of Anatomic Pathology information supports the development of computer-based decision support as well as secondary use of Anatomic Pathology information for research or public health.

Pathology information systems have been using SNOMED II for many years, and in most cases, they are in a migration process to SNOMED CT. COST Action IC0604 (EURO-TELEPATH) has considered terminology normalization one of its strategic objectives. This paper reviews the use of SNOMED CT in healthcare, with a special focus in pathology. Nowadays, SNOMED CT is mainly used for concept search and coding of clinical data. Some ontological errors found in SNOMED CT are described. The Integrating the Healthcare Enterprise (IHE) initiative has fostered the use of SNOMED CT, also in Pathology, as recommended in the Supplement Anatomic Pathology Structured Reports of the IHE Anatomic Pathology Technical Framework. Rule governing concept post-coordination is also described. Some recent initiatives are trying to define a SNOMED CT subset for Pathology. The Spanish Society of Pathology has defined a subset for specimens and procedures in Pathology. Regarding diagnosis coding, the morphological abnormality sub-hierarchy of SNOMED CT need to be significantly extended and improved to become useful for pathologists. A consensus is needed to encode pathology reports with the adequate hierarchies and concepts. This will make the implementation of pathology structured reports more feasible.

The development of small molecule inhibitors of the growth factor receptors and discovery of somatic mutations of the thyrosine kinase domain resulted in new paradigms for the cancers therapy. Digital microscopy is an important tool for surgical pathologists. The achievements in the digital pathology field have modified the workflow of pathomorphology labs, enhanced the pathologists' role in the diagnostics and increased their contribution to the personalized targeted medicine. Digital image analysis is now available in a variety of platforms to improve quantification performance of diagnostic pathology. The authors describe the state of digital microscopy as it applies to the field of quantitative immunohistochemistry of biomarkers related to the clinical personalized targeted therapy of breast cancer, non-small lung cancer and colorectal cancer: HER-2, EGFR, KRAS and BRAF genes. The information is derived from the experience of the authors and review of the literature.

In the current practice of pathology, the evaluation of immunohistochemical (IHC) markers represents an essential tool. The manual quantification of these markers is still laborious and subjective, and the use of computerized systems for digital image (DI) analysis has not yet resolved the problems of nuclear aggregates (clusters). Furthermore, the volume of DI storage continues to be an important problem in computer-assisted pathology. In the present study we have developed an automated procedure to quantify IHC nuclear markers in DI with a high level of clusters. Furthermore the effects of JPEG compression in the image analysis were evaluated. The results indicated that there was an agreement with the results of both methods (automated vs. manual) in almost 90% of the analyzed images. On the other hand, automated count differences increase as the compression level increase, but only in images with a high number of stained nuclei (>nuclei/image) or with high area cluster (>25μm²). Some corrector factors were developed in order to correct this count differences. In conclusion, the proposed automated procedure is an objective, faster than manual counting and reproducible method that has more than 90% of similarity with manual count. Moreover, the results demonstrate that with correction factors, it is possible to carry out unbiased automated quantifications on IHC nuclear markers in compressed DIs.

The paper describes a content-based image retrieval (CBIR) system with relevance feedback (RF). Instead of standard relevance feedback procedure, an adaptive clustering of image database (ACID) according to particular subjective needs is introduced in our system. Images labeled by the user as relevant are collected in clusters, and their representative members are used in further searching procedure instead of all images contained in the database. By this way, some history of previous retrieving is embedded into a searching process enabling faster and more subjective retrieval. Moreover, clusters are adaptively updated after each retrieving session, following actual user's needs. The efficiency of the proposed ACID system is tested with images from Corel and MIT datasets.

Background: Whole Slide Imagers or digital slide scanners have developed very rapidly in the last 8 years and went through three generations. Third generation instruments have just reached the market which have the stability and throughput to be used for routine clinical work. We describe in this article the technical background and reasoning behind engineering decisions we made during the development of 3DHISTECH's 3rd generation combined brightfield and fluorescent scanner.

Materials and methods: The Pannoramic 250 FLASH utilizes Plan-Apochromat 20x and 40x objectives, a 2 megapixel 3CCD camera for brightfield and a monochrome scientific CMOS camera for fluorescent scanning. A solid state light engine for fluorescent and a strobe light for bright field illumination are used.

Results: The system can scan 1cm² including focusing at 45x resolution in 1 minute. It can scan a well stained DAPI, FITC, TRIC, 1cm² fluorescent slide in 11 minutes. It can load and scan 250 slides in walk away mode.

Conclusion: Using the latest camera technology and electronics, state of the art computer and standard microscope optical components high throughput high quality whole slide imaging is feasible and is sufficient for most of the routine diagnostic work. Extended depth of field and Z-stack scanning is possible with the use of area scan technology.

Ubiquitous personalized health services including ePathology require comprehensive, but trusted interoperability. Contrary to regulated traditional health services with pre-defined policies, the solutions enabled by mobile, pervasive and autonomous technology have to follow dynamic policies reflecting the customers changing health services needs, expectations and wishes as well as contextual and environmental conditions. The paper introduces an advanced approach to trustworthy architecture-centric, policy-driven pHealth solutions. To some details, it also addresses security and privacy ontologies to represent the required policies.

Grid technology has enabled clustering and access to, and interaction among, a wide variety of geographically distributed resources such as supercomputers, storage systems, data sources, instruments as well as special devices and services, realizing network-centric operations. Their main applications include large scale computational and data intensive problems in science and engineering. Grids are likely to have a deep impact on health related applications. Moreover, they seem to be suitable for tissue-based diagnosis. They offer a powerful tool to deal with current challenges in many biomedical domains involving complex anatomical and physiological modeling of structures from images or large image databases assembling and analysis. This chapter analyzes the general structures and functions of a Grid environment implemented for tissue-based diagnosis on digital images. Moreover, it presents a Grid middleware implemented by the authors for diagnostic pathology applications. The chapter is a review of the work done as part of the European COST project EUROTELEPATH.