Bridging the Clinical-Computational Transparency Gap in Digital Pathology.

CONTEXT.­: Computational pathology combines clinical pathology with computational analysis, aiming to enhance diagnostic capabilities and improve clinical productivity. However, communication barriers between pathologists and developers often hinder the full realization of this potential. OBJECTIVE.­: To propose a standardized framework that improves mutual understanding of clinical objectives and computational methodologies. The goal is to enhance the development and application of computer-aided diagnostic (CAD) tools. DESIGN.­: The article suggests pivotal roles for pathologists and computer scientists in the CAD development process. It calls for increased understanding of computational terminologies, processes, and limitations among pathologists. Similarly, it argues that computer scientists should better comprehend the true use cases of the developed algorithms to avoid clinically meaningless metrics. RESULTS.­: CAD tools improve pathology practice significantly. Some tools have even received US Food and Drug Administration approval. However, improved understanding of machine learning models among pathologists is essential to prevent misuse and misinterpretation. There is also a need for a more accurate representation of the algorithms' performance compared to that of pathologists. CONCLUSIONS.­: A comprehensive understanding of computational and clinical paradigms is crucial for overcoming the translational gap in computational pathology. This mutual comprehension will improve patient care through more accurate and efficient disease diagnosis.

Recommendations for Performance Evaluation of Machine Learning in Pathology: A Concept Paper From the College of American Pathologists.

CONTEXT.­: Machine learning applications in the pathology clinical domain are emerging rapidly. As decision support systems continue to mature, laboratories will increasingly need guidance to evaluate their performance in clinical practice. Currently there are no formal guidelines to assist pathology laboratories in verification and/or validation of such systems. These recommendations are being proposed for the evaluation of machine learning systems in the clinical practice of pathology. OBJECTIVE.­: To propose recommendations for performance evaluation of in vitro diagnostic tests on patient samples that incorporate machine learning as part of the preanalytical, analytical, or postanalytical phases of the laboratory workflow. Topics described include considerations for machine learning model evaluation including risk assessment, predeployment requirements, data sourcing and curation, verification and validation, change control management, human-computer interaction, practitioner training, and competency evaluation. DATA SOURCES.­: An expert panel performed a review of the literature, Clinical and Laboratory Standards Institute guidance, and laboratory and government regulatory frameworks. CONCLUSIONS.­: Review of the literature and existing documents enabled the development of proposed recommendations. This white paper pertains to performance evaluation of machine learning systems intended to be implemented for clinical patient testing. Further studies with real-world clinical data are encouraged to support these proposed recommendations. Performance evaluation of machine learning models is critical to verification and/or validation of in vitro diagnostic tests using machine learning intended for clinical practice.

Artificial intelligence and digital pathology: clinical promise and deployment considerations.

Artificial intelligence (AI) presents an opportunity in anatomic pathology to provide quantitative objective support to a traditionally subjective discipline, thereby enhancing clinical workflows and enriching diagnostic capabilities. AI requires access to digitized pathology materials, which, at present, are most commonly generated from the glass slide using whole-slide imaging. Models are developed collaboratively or sourced externally, and best practices suggest validation with internal datasets most closely resembling the data expected in practice. Although an array of AI models that provide operational support for pathology practices or improve diagnostic quality and capabilities has been described, most of them can be categorized into one or more discrete types. However, their function in the pathology workflow can vary, as a single algorithm may be appropriate for screening and triage, diagnostic assistance, virtual second opinion, or other uses depending on how it is implemented and validated. Despite the clinical promise of AI, the barriers to adoption have been numerous, to which inclusion of new stakeholders and expansion of reimbursement opportunities may be among the most impactful solutions.

High-throughput whole-slide scanning to enable large-scale data repository building.

Digital pathology and artificial intelligence (AI) rely on digitization of patient material as a necessary first step. AI development benefits from large sample sizes and diverse cohorts, and therefore efforts to digitize glass slides must meet these needs in an efficient and cost-effective manner. Technical innovation in whole-slide imaging has enabled high-throughput slide scanning through the coordinated increase in scanner capacity, speed, and automation. Combining these hardware innovations with automated informatics approaches has enabled more efficient workflows and the opportunity to provide higher-quality imaging data using fewer personnel. Here we review several practical considerations for deploying high-throughput scanning and we present strategies to increase efficiency with a focus on quality. Finally, we review remaining challenges and issue a call to vendors to innovate in the areas of automation and quality control in order to make high-throughput scanning realizable to laboratories with limited resources. © 2022 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of The Pathological Society of Great Britain and Ireland.

Dissecting the Business Case for Adoption and Implementation of Digital Pathology: A White Paper from the Digital Pathology Association.

We believe the switch to a digital pathology (DP) workflow is imminent and it is essential to understand the economic implications of conversion. Many aspects of the adoption of DP will be disruptive and have a direct financial impact, both in short term costs, such as investment in equipment and personnel, and long term revenue potential, such as improved productivity and novel tests. The focus of this whitepaper is to educate pathologists, laboratorians and other stakeholders about the business and monetary considerations of converting to a digital pathology workflow. The components of a DP business plan will be thoroughly summarized, and guidance will be provided on how to build a case for adoption and implementation as well as a roadmap for transitioning from an analog to a digital pathology workflow in various laboratory settings. It is important to clarify that this publication is not intended to list prices although some financials will be mentioned as examples. The authors encourage readers who are evaluating conversion to a DP workflow to use this paper as a foundational guide for conducting a thorough and complete assessment while incorporating in current market pricing. Contributors to this paper analyzed peer-reviewed literature and data collected from various institutions, some of which are mentioned. Digital pathology will change the way we practice through facilitating patient access to expert pathology services and enabling image analysis tools and assays to aid in diagnosis, prognosis, risk stratification and therapeutic selection. Together, they will result in the delivery of valuable information from which to make better decisions and improve the health of patients.

Quantitative Image Analysis for Tissue Biomarker Use: A White Paper From the Digital Pathology Association.

Tissue biomarkers have been of increasing utility for scientific research, diagnosing disease, and treatment response prediction. There has been a steady shift away from qualitative assessment toward providing more quantitative scores for these biomarkers. The application of quantitative image analysis has thus become an indispensable tool for in-depth tissue biomarker interrogation in these contexts. This white paper reviews current technologies being employed for quantitative image analysis, their application and pitfalls, regulatory framework demands, and guidelines established for promoting their safe adoption in clinical practice.

Continuing Undergraduate Pathology Medical Education in the Coronavirus Disease 2019 (COVID-19) Global Pandemic: The Johns Hopkins Virtual Surgical Pathology Clinical Elective.

CONTEXT.­: In the early months of the response to the coronavirus disease 2019 (COVID-19) pandemic, the Johns Hopkins University School of Medicine (JHUSOM) (Baltimore, Maryland) leadership reached out to faculty to develop and implement virtual clinical clerkships after all in-person medical student clinical experiences were suspended. OBJECTIVE.­: To develop and implement a digital slide-based virtual surgical pathology (VSP) clinical elective to meet the demand for meaningful and robust virtual clinical electives in response to the temporary suspension of in-person clinical rotations at JHUSOM. DESIGN.­: The VSP elective was modeled after the in-person surgical pathology elective to include virtual previewing and sign-out with standardized cases supplemented by synchronous and asynchronous pathology educational content. RESULTS.­: Validation of existing Web communications technology and slide-scanning systems was performed by feasibility testing. Curriculum development included drafting of course objectives and syllabus, Blackboard course site design, electronic-lecture creation, communications with JHUSOM leadership, scheduling, and slide curation. Subjectively, the weekly schedule averaged 35 to 40 hours of asynchronous, synchronous, and independent content, approximately 10 to 11 hours of which were synchronous. As of February 2021, VSP has hosted 35 JHUSOM and 8 non-JHUSOM students, who have provided positive subjective and objective course feedback. CONCLUSIONS.­: The Johns Hopkins VSP elective provided meaningful clinical experience to 43 students in a time of immense online education need. Added benefits of implementing VSP included increased medical student exposure to pathology as a medical specialty and demonstration of how digital slides have the potential to improve standardization of the pathology clerkship curriculum.

Computational pathology definitions, best practices, and recommendations for regulatory guidance: a white paper from the Digital Pathology Association.

In this white paper, experts from the Digital Pathology Association (DPA) define terminology and concepts in the emerging field of computational pathology, with a focus on its application to histology images analyzed together with their associated patient data to extract information. This review offers a historical perspective and describes the potential clinical benefits from research and applications in this field, as well as significant obstacles to adoption. Best practices for implementing computational pathology workflows are presented. These include infrastructure considerations, acquisition of training data, quality assessments, as well as regulatory, ethical, and cyber-security concerns. Recommendations are provided for regulators, vendors, and computational pathology practitioners in order to facilitate progress in the field. © 2019 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of Pathological Society of Great Britain and Ireland.

Introduction to Digital Image Analysis in Whole-slide Imaging: A White Paper from the Digital Pathology Association.

The advent of whole-slide imaging in digital pathology has brought about the advancement of computer-aided examination of tissue via digital image analysis. Digitized slides can now be easily annotated and analyzed via a variety of algorithms. This study reviews the fundamentals of tissue image analysis and aims to provide pathologists with basic information regarding the features, applications, and general workflow of these new tools. The review gives an overview of the basic categories of software solutions available, potential analysis strategies, technical considerations, and general algorithm readouts. Advantages and limitations of tissue image analysis are discussed, and emerging concepts, such as artificial intelligence and machine learning, are introduced. Finally, examples of how digital image analysis tools are currently being used in diagnostic laboratories, translational research, and drug development are discussed.

Video compression to support the expansion of whole-slide imaging into cytology.

Digital screening and diagnosis from cytology slides can be aided by capturing multiple focal planes. However, using conventional methods, the large file sizes of high-resolution whole-slide images increase linearly with the number of focal planes acquired, leading to significant data storage and bandwidth requirements for the efficient storage and transfer of cytology virtual slides. We investigated whether a sequence of focal planes contained sufficient redundancy to efficiently compress virtual slides across focal planes by applying a commonly available video compression standard, high-efficiency video coding (HEVC). By developing an adaptive algorithm that applied compression to achieve a target image quality, we found that the compression ratio of HEVC exceeded that obtained using JPEG and JPEG2000 compression while maintaining a comparable level of image quality. These results suggest an alternative method for the efficient storage and transfer of whole-slide images that contain multiple focal planes, expanding the utility of this rapidly evolving imaging technology into cytology.

A Practical Guide to Whole Slide Imaging: A White Paper From the Digital Pathology Association.

CONTEXT.­: Whole slide imaging (WSI) represents a paradigm shift in pathology, serving as a necessary first step for a wide array of digital tools to enter the field. Its basic function is to digitize glass slides, but its impact on pathology workflows, reproducibility, dissemination of educational material, expansion of service to underprivileged areas, and intrainstitutional and interinstitutional collaboration exemplifies a significant innovative movement with far-reaching effects. Although the benefits of WSI to pathology practices, academic centers, and research institutions are many, the complexities of implementation remain an obstacle to widespread adoption. In the wake of the first regulatory clearance of WSI for primary diagnosis in the United States, some barriers to adoption have fallen. Nevertheless, implementation of WSI remains a difficult prospect for many institutions, especially those with stakeholders unfamiliar with the technologies necessary to implement a system or who cannot effectively communicate to executive leadership and sponsors the benefits of a technology that may lack clear and immediate reimbursement opportunity. OBJECTIVES.­: To present an overview of WSI technology-present and future-and to demonstrate several immediate applications of WSI that support pathology practice, medical education, research, and collaboration. DATA SOURCES.­: Peer-reviewed literature was reviewed by pathologists, scientists, and technologists who have practical knowledge of and experience with WSI. CONCLUSIONS.­: Implementation of WSI is a multifaceted and inherently multidisciplinary endeavor requiring contributions from pathologists, technologists, and executive leadership. Improved understanding of the current challenges to implementation, as well as the benefits and successes of the technology, can help prospective users identify the best path for success.

BCL-2 expression aids in the immunohistochemical prediction of the Oncotype DX breast cancer recurrence score.

BACKGROUND: The development of molecular techniques to estimate the risk of breast cancer recurrence has been a significant addition to the suite of tools available to pathologists and breast oncologists. It has previously been shown that immunohistochemistry can provide a surrogate measure of tumor recurrence risk, effectively providing a less expensive and more rapid estimate of risk without the need for send-out. However, concordance between gene expression-based and immunohistochemistry-based approaches has been modest, making it difficult to determine when one approach can serve as an adequate substitute for the other. We investigated whether immunohistochemistry-based methods can be augmented to provide a useful therapeutic indicator of risk. METHODS: We studied whether the Oncotype DX breast cancer recurrence score can be predicted from routinely acquired immunohistochemistry of breast tumor histology. We examined the effects of two modifications to conventional scoring measures based on ER, PR, Ki-67, and Her2 expression. First, we tested a mathematical transformation that produces a more diagnostic-relevant representation of the staining attributes of these markers. Second, we considered the expression of BCL-2, a complex involved in regulating apoptosis, as an additional prognostic marker. RESULTS: We found that the mathematical transformation improved concordance rates over the conventional scoring model. By establishing a measure of prediction certainty, we discovered that the difference in concordance between methods was even greater among the most certain cases in the sample, demonstrating the utility of an accompanying measure of prediction certainty. Including BCL-2 expression in the scoring model increased the number of breast cancer cases in the cohort that were considered high certainty, effectively expanding the applicability of this technique to a greater proportion of patients. CONCLUSIONS: Our results demonstrate an improvement in concordance between immunohistochemistry-based and gene expression-based methods to predict breast cancer recurrence risk following two simple modifications to the conventional scoring model.

Estimation of Fine-Scale Histologic Features at Low Magnification.

CONTEXT.­: Whole-slide imaging has ushered in a new era of technology that has fostered the use of computational image analysis for diagnostic support and has begun to transfer the act of analyzing a slide to computer monitors. Due to the overwhelming amount of detail available in whole-slide images, analytic procedures-whether computational or visual-often operate at magnifications lower than the magnification at which the image was acquired. As a result, a corresponding reduction in image resolution occurs. It is unclear how much information is lost when magnification is reduced, and whether the rich color attributes of histologic slides can aid in reconstructing some of that information. OBJECTIVE.­: To examine the correspondence between the color and spatial properties of whole-slide images to elucidate the impact of resolution reduction on the histologic attributes of the slide. DESIGN.­: We simulated image resolution reduction and modeled its effect on classification of the underlying histologic structure. By harnessing measured histologic features and the intrinsic spatial relationships between histologic structures, we developed a predictive model to estimate the histologic composition of tissue in a manner that exceeds the resolution of the image. RESULTS.­: Reduction in resolution resulted in a significant loss of the ability to accurately characterize histologic components at magnifications less than ×10. By utilizing pixel color, this ability was improved at all magnifications. CONCLUSIONS.­: Multiscale analysis of histologic images requires an adequate understanding of the limitations imposed by image resolution. Our findings suggest that some of these limitations may be overcome with computational modeling.

Laboratory Computer Performance in a Digital Pathology Environment: Outcomes from a Single Institution.

BACKGROUND: In an effort to provide improved user experience and system reliability at a moderate cost, our department embarked on targeted upgrades of a total of 87 computers over a period of 3 years. Upgrades came in three forms: (i) replacement of the computer with newer architecture, (ii) replacement of the computer's hard drive with a solid-state drive (SSD), or (iii) replacement of the computer with newer architecture and a SSD. METHODS: We measured the impact of each form of upgrade on a set of pathology-relevant tasks that fell into three categories: standard use, whole-slide navigation, and whole-slide analysis. We used time to completion of a task as the primary variable of interest. RESULTS: We found that for most tasks, the SSD upgrade had a greater impact than the upgrade in architecture. This effect was especially prominent for whole-slide viewing, likely due to the way in which most whole-slide viewers cached image tiles. However, other tasks, such as whole-slide image analysis, often relied less on disk input or output and were instead more sensitive to the computer architecture. CONCLUSIONS: Based on our experience, we suggest that SSD upgrades are viewed in some settings as a viable alternative to complete computer replacement and recommend that computer replacements in a digital pathology setting are accompanied by an upgrade to SSDs.

Contextual modulation revealed by optical imaging exhibits figural asymmetry in macaque V1 and V2.

Neurons in early visual cortical areas are influenced by stimuli presented well beyond the confines of their classical receptive fields, endowing them with the ability to encode fine-scale features while also having access to the global context of the visual scene. This property can potentially define a role for the early visual cortex to contribute to a number of important visual functions, such as surface segmentation and figure-ground segregation. It is unknown how extraclassical response properties conform to the functional architecture of the visual cortex, given the high degree of functional specialization in areas V1 and V2. We examined the spatial relationships of contextual activations in macaque V1 and V2 with intrinsic signal optical imaging. Using figure-ground stimulus configurations defined by orientation or motion, we found that extraclassical modulation is restricted to the cortical representations of the figural component of the stimulus. These modulations were positive in sign, suggesting a relative enhancement in neuronal activity that may reflect an excitatory influence. Orientation and motion cues produced similar patterns of activation that traversed the functional subdivisions of V2. The asymmetrical nature of the enhancement demonstrated the capacity for visual cortical areas as early as V1 to contribute to figure-ground segregation, and the results suggest that this information can be extracted from the population activity constrained only by retinotopy, and not the underlying functional organization.

An alternative reference space for H&E color normalization.

Digital imaging of H&E stained slides has enabled the application of image processing to support pathology workflows. Potential applications include computer-aided diagnostics, advanced quantification tools, and innovative visualization platforms. However, the intrinsic variability of biological tissue and the vast differences in tissue preparation protocols often lead to significant image variability that can hamper the effectiveness of these computational tools. We developed an alternative representation for H&E images that operates within a space that is more amenable to many of these image processing tools. The algorithm to derive this representation operates by exploiting the correlation between color and the spatial properties of the biological structures present in most H&E images. In this way, images are transformed into a structure-centric space in which images are segregated into tissue structure channels. We demonstrate that this framework can be extended to achieve color normalization, effectively reducing inter-slide variability.

Cue combination encoding via contextual modulation of V1 and V2 neurons.

Neurons in early visual cortical areas encode the local properties of a stimulus in a number of different feature dimensions such as color, orientation, and motion. It has been shown, however, that stimuli presented well beyond the confines of the classical receptive field can augment these responses in a way that emphasizes these local attributes within the greater context of the visual scene. This mechanism imparts global information to cells that are otherwise considered local feature detectors and can potentially serve as an important foundation for surface segmentation, texture representation, and figure-ground segregation. The role of early visual cortex toward these functions remains somewhat of an enigma, as it is unclear how surface segmentation cues are integrated from multiple feature dimensions. We examined the impact of orientation- and motion-defined surface segmentation cues in V1 and V2 neurons using a stimulus in which the two features are completely separable. We find that, although some cells are modulated in a cue-invariant manner, many cells are influenced by only one cue or the other. Furthermore, cells that are modulated by both cues tend to be more strongly affected when both cues are presented together than when presented individually. These results demonstrate two mechanisms by which cue combinations can enhance salience. We find that feature-specific populations are more frequently encountered in V1, while cue additivity is more prominent in V2. These results highlight how two strongly interconnected areas at different stages in the cortical hierarchy can potentially contribute to scene segmentation.

An optimized color transformation for the analysis of digital images of hematoxylin & eosin stained slides.

Hematoxylin and eosin (H&E) staining is ubiquitous in pathology practice and research. As digital pathology has evolved, the reliance of quantitative methods that make use of H&E images has similarly expanded. For example, cell counting and nuclear morphometry rely on the accurate demarcation of nuclei from other structures and each other. One of the major obstacles to quantitative analysis of H&E images is the high degree of variability observed between different samples and different laboratories. In an effort to characterize this variability, as well as to provide a substrate that can potentially mitigate this factor in quantitative image analysis, we developed a technique to project H&E images into an optimized space more appropriate for many image analysis procedures. We used a decision tree-based support vector machine learning algorithm to classify 44 H&E stained whole slide images of resected breast tumors according to the histological structures that are present. This procedure takes an H&E image as an input and produces a classification map of the image that predicts the likelihood of a pixel belonging to any one of a set of user-defined structures (e.g., cytoplasm, stroma). By reducing these maps into their constituent pixels in color space, an optimal reference vector is obtained for each structure, which identifies the color attributes that maximally distinguish one structure from other elements in the image. We show that tissue structures can be identified using this semi-automated technique. By comparing structure centroids across different images, we obtained a quantitative depiction of H&E variability for each structure. This measurement can potentially be utilized in the laboratory to help calibrate daily staining or identify troublesome slides. Moreover, by aligning reference vectors derived from this technique, images can be transformed in a way that standardizes their color properties and makes them more amenable to image processing.

Lymph Node Metastasis Status in Breast Carcinoma Can Be Predicted via Image Analysis of Tumor Histology.

OBJECTIVE: To develop a method whereby axillary lymph node (ALN) metastasis can be predicted without ALN dissection, via computational image analysis of routinely acquired tumor histology. STUDY DESIGN: We employed digital image processing to stratify patients based on the histological attributes of the primary tumor. We extracted image features that capture the nuclear and architectural properties of the specimen. We then used a novel machine learning algorithm to transform image features into a scalar score that provided not only a metastasis prediction but also the certainty of classification. RESULTS: We applied this procedure to 101 patients with a ground truth established by histological examination of the lymph nodes and found that 68.3% of the cohort could be classified, exhibiting a correct prediction rate of 88.4%. CONCLUSION: These results demonstrate a technique that potentially can be used to supplant existing surgical methods to determine ALN metastasis status, thereby reducing patient morbidity associated with over-treatment.