Revolutionizing Digital Pathology With the Power of Generative Artificial Intelligence and Foundation Models.

Digital pathology has transformed the traditional pathology practice of analyzing tissue under a microscope into a computer vision workflow. Whole-slide imaging allows pathologists to view and analyze microscopic images on a computer monitor, enabling computational pathology. By leveraging artificial intelligence (AI) and machine learning (ML), computational pathology has emerged as a promising field in recent years. Recently, task-specific AI/ML (eg, convolutional neural networks) has risen to the forefront, achieving above-human performance in many image-processing and computer vision tasks. The performance of task-specific AI/ML models depends on the availability of many annotated training datasets, which presents a rate-limiting factor for AI/ML development in pathology. Task-specific AI/ML models cannot benefit from multimodal data and lack generalization, eg, the AI models often struggle to generalize to new datasets or unseen variations in image acquisition, staining techniques, or tissue types. The 2020s are witnessing the rise of foundation models and generative AI. A foundation model is a large AI model trained using sizable data, which is later adapted (or fine-tuned) to perform different tasks using a modest amount of task-specific annotated data. These AI models provide in-context learning, can self-correct mistakes, and promptly adjust to user feedback. In this review, we provide a brief overview of recent advances in computational pathology enabled by task-specific AI, their challenges and limitations, and then introduce various foundation models. We propose to create a pathology-specific generative AI based on multimodal foundation models and present its potentially transformative role in digital pathology. We describe different use cases, delineating how it could serve as an expert companion of pathologists and help them efficiently and objectively perform routine laboratory tasks, including quantifying image analysis, generating pathology reports, diagnosis, and prognosis. We also outline the potential role that foundation models and generative AI can play in standardizing the pathology laboratory workflow, education, and training.

External Quality Assessment Programs in the US with an emphasis on urinary sediment testing: the College of American Pathologists experience.

The College of American Pathologists (CAP) has maintained the highest standards for laboratory medicine through education, evaluation, and certification. One form of External Quality Assurance - proficiency testing (PT) - is the centerpiece of that mission. Over 500 medical and scientific experts oversee CAP PT programs which include more than 600 tests performed by 22,000 laboratories in over 100 countries. It is the most comprehensive laboratory peer-review comparison program in the world. The CAP offers four urine sediment PT products tailored to the needs of different laboratories. Each includes three or four digital images, shipped twice a year. The program is overseen by the Hematology and Clinical Microscopy Resource Committee. Images are graded if there is 80% or greater consensus of either referee or participant laboratories. Failing laboratories must analyze the reasons for the failure, report the results, and initiate corrective action. Over the years, there has been a progressive decline in the number of errors, demonstrating that education and regulatory oversight are major contributors to improved PT performance and, by extension, patient care. The PT urine sediment image databank is a unique resource, representing the consensus of many laboratories. Participant and referee responses identify which morphologic variants are unambiguous and which are more difficult to classify. The PT challenges include discussions of disease pathophysiology and key morphologic features. This teaching component is what helps to set the CAP's program apart. The discussions formed the basis for the Color Atlas of Urinary Sediment published by the CAP in 2010.

A Pilot Validation Study Comparing Fluorescence-Imitating Brightfield Imaging, A Slide-Free Imaging Method, With Standard Formalin-Fixed, Paraffin-Embedded Hematoxylin-Eosin-Stained Tissue Section Histology for Primary Surgical Pathology Diagnosis.

CONTEXT.­: Digital pathology using whole slide images has been recently approved to support primary diagnosis in clinical surgical pathology practices. Here we describe a novel imaging method, fluorescence-imitating brightfield imaging, that can capture the surface of fresh tissue without requiring prior fixation, paraffin embedding, tissue sectioning, or staining. OBJECTIVE.­: To compare the ability of pathologists to evaluate direct-to-digital images with standard pathology preparations. DESIGN.­: One hundred surgical pathology samples were obtained. Samples were first digitally imaged, then processed for standard histologic examination on 4-µm hematoxylin-eosin-stained sections and digitally scanned. The resulting digital images from both digital and standard scan sets were viewed by each of 4 reading pathologists. The data set consisted of 100 reference diagnoses and 800 study pathologist reads. Each study read was compared to the reference diagnosis, and also compared to that reader's diagnosis across both modalities. RESULTS.­: The overall agreement rate, across 800 reads, was 97.9%. This consisted of 400 digital reads at 97.0% versus reference and 400 standard reads versus reference at 98.8%. Minor discordances (defined as alternative diagnoses without clinical treatment or outcome implications) were 6.1% overall, 7.2% for digital, and 5.0% for standard. CONCLUSIONS.­: Pathologists can provide accurate diagnoses from fluorescence-imitating brightfield imaging slide-free images. Concordance and discordance rates are similar to published rates for comparisons of whole slide imaging to standard light microscopy of glass slides for primary diagnosis. It may be possible, therefore, to develop a slide-free, nondestructive approach for primary pathology diagnosis.

Harnessing the Power of Generative Artificial Intelligence in Pathology Education.

CONTEXT.­: Generative artificial intelligence (AI) technologies are rapidly transforming numerous fields, including pathology, and hold significant potential to revolutionize educational approaches. OBJECTIVE.­: To explore the application of generative AI, particularly large language models and multimodal tools, for enhancing pathology education. We describe their potential to create personalized learning experiences, streamline content development, expand access to educational resources, and support both learners and educators throughout the training and practice continuum. DATA SOURCES.­: We draw on insights from existing literature on AI in education and the collective expertise of the coauthors within this rapidly evolving field. Case studies highlight practical applications of large language models, demonstrating both the potential benefits and unique challenges associated with implementing these technologies in pathology education. CONCLUSIONS.­: Generative AI presents a powerful tool kit for enriching pathology education, offering opportunities for greater engagement, accessibility, and personalization. Careful consideration of ethical implications, potential risks, and appropriate mitigation strategies is essential for the responsible and effective integration of these technologies. Future success lies in fostering collaborative development between AI experts and medical educators, prioritizing ongoing human oversight and transparency to ensure that generative AI augments, rather than supplants, the vital role of educators in pathology training and practice.

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.

Pathology Education Powered by Virtual and Digital Transformation: Now and the Future.

CONTEXT.­: Myriad forces are changing teaching and learning strategies throughout all stages and types of pathology education. Pathology educators and learners face the challenge of adapting to and adopting new methods and tools. The digital pathology transformation and the associated educational ecosystem are major factors in this setting of change. OBJECTIVE.­: To identify and collect resources, tools, and examples of educational innovations involving digital pathology that are valuable to pathology learners and teachers at each phase of professional development. DATA SOURCES.­: Sources were a literature review and the personal experience of authors and educators. CONCLUSIONS.­: High-quality digital pathology tools and resources have permeated all the major niches within anatomic pathology and are increasingly well applied to clinical pathology for learners at all levels. Coupled with other virtual tools, the training landscape in pathology is highly enriched and much more accessible than in the past. Digital pathology is well suited to the demands of peer-to-peer education, such as in the introduction of new testing, grading, or other standardized practices. We found that digital pathology was well adapted to apply our current understanding of optimal teaching strategies and was effective at the undergraduate, graduate, postgraduate, and peer-to-peer levels. We curated and tabulated many existing resources within some segments of pathology. We identified several best practices for each training or educational stage based on current materials and proposed high-priority areas for potential future development.

Normal Routine Spirometry Can Mask COPD/Emphysema in Symptomatic Smokers.

BACKGROUND: Recent studies have emphasized the difficulty of early detection of chronic obstructive pulmonary disease (COPD) in symptomatic smokers with normal routine spirometry. This includes post-bronchodilator normal forced expiratory volume in 1 second (FEV1)(L)≥80% predicted, forced vital capacity (FVC)(L)≥80% predicted, and FEV1/FVC ≥70% or greater than age corrected lower limit of normal (LLN). However, in COPD the pathologic site of small airway obstruction and emphysema begins in the small peripheral airways ≤2 mm id which normally contribute <20% of total airway resistance. METHODS: Expiratory airflow at high and low lung volumes post-bronchodilator were measured and correlated with lung computed tomography (CT) and lung pathology (6 patients) in 16 symptomatic, treated smokers, and all with normal routine spirometry. RESULTS: Despite normal routine spirometry, all16 patients had isolated, abnormal forced expiratory flow at 75% of FVC (FEF75) using data from Knudson et al, Hankinson et al NHAMES III, and Quanjer et al and the Global Lung Function Initiative. This reflects isolated detection of small airways obstruction and/or emphysema. Measuring airflow at FEF50 detected only 8 of 16 patients, maximal expiratory flow at 25%-75% of FVC (MEF25-75) only 4 of 16, residual volume (RV) 4 of 16, and RV to total lung capacity ratio only 2 of 16. There was excellent correlation between limited lung pathology and lung CT for absence of emphysema. CONCLUSION: This study confirms our earlier observations that detection of small airways obstruction and/or emphysema, in symptomatic smokers with normal routine spirometry, requires analysis of expiratory airflow at low lung volumes, including FEF75. Dependence upon normal routine spirometry may result in clinical and physiologic delay in the diagnosis and treatment in symptomatic smokers with emphysema and small airways obstruction.

Digital Whole Slide Imaging Compared With Light Microscopy for Primary Diagnosis in Surgical Pathology.

CONTEXT.­: The adoption of digital capture of pathology slides as whole slide images (WSI) for educational and research applications has proven utility. OBJECTIVE.­: To compare pathologists' primary diagnoses derived from WSI versus the standard microscope. Because WSIs differ in format and method of observation compared with the current standard glass slide microscopy, this study is critical to potential clinical adoption of digital pathology. DESIGN.­: The study enrolled a total of 2045 cases enriched for more difficult diagnostic categories and represented as 5849 slides were curated and provided for diagnosis by a team of 19 reading pathologists separately as WSI or as glass slides viewed by light microscope. Cases were reviewed by each pathologist in both modalities in randomized order with a minimum 31-day washout between modality reads for each case. Each diagnosis was compared with the original clinical reference diagnosis by an independent central adjudication review. RESULTS.­: The overall major discrepancy rates were 3.64% for WSI review and 3.20% for manual slide review diagnosis methods, a difference of 0.44% (95% CI, -0.15 to 1.03). The time to review a case averaged 5.20 minutes for WSI and 4.95 minutes for glass slides. There was no specific subset of diagnostic category that showed higher rates of modality-specific discrepancy, though some categories showed greater discrepancy than others in both modalities. CONCLUSIONS.­: WSIs are noninferior to traditional glass slides for primary diagnosis in anatomic pathology.

Further Studies of Unsuspected Emphysema in Nonsmoking Patients With Asthma With Persistent Expiratory Airflow Obstruction.

BACKGROUND: Previously, we and other investigators have described reversible loss of lung elastic recoil in patients with acute and persistent, moderate-to-severe, chronic, treated asthma who never smoked, and its adverse effect on maximal expiratory airflow. In four consecutive autopsies, we reported the pathophysiologic mechanism(s) has been unsuspected mild, diffuse, middle and upper lobe centrilobular emphysema. METHODS: We performed prospective studies (5 to 22 years) in 25 patients (12 female) with chronic asthma, age 55 ± 15 years, who never smoked, with persistent moderate-to-severe expiratory obstruction. Studies included measuring blood eosinophils, IgE, total exhaled nitric oxide (NO), central airway NO flux, peripheral airway/alveolar NO concentration, impulse oscillometry, heliox curves, lung elastic recoil, and high-resolution thin-section (1 mm) lung CT imaging at full inspiration with voxel quantification. RESULTS: In 25 patients with stable asthma with varying type 2 phenotype, after 270 µg of aerosolized albuterol sulfate had been administered with a metered dose inhaler with space chamber, FVC was 3.1 ± 1.0 L (83% ± 13% predicted) (mean ± SD), FEV1 was 1.8 ± 0.6 L (59% ± 11%), the FEV1/FVC ratio was 59% ± 10%, and the ratio of single-breath diffusing capacity of the lung for carbon monoxide to alveolar volume was 4.8 ± 1.1 mL/min/mm Hg/L (120% ± 26%). All 25 patients with asthma had loss of static lung elastic recoil pressure, which contributed equally to decreased intrinsic airway conductance in limiting expiratory airflow. Lung CT scanning detected none or mild emphysema. In all four autopsied asthmatic lungs previously reported and one unreported explanted lung, microscopy revealed unsuspected mild, diffuse centrilobular emphysema in the upper and middle lung fields, and asthma-related remodeling in airways. In eight cases, during asthma remission, there were increases in measured static lung elastic recoil pressure-calculated intrinsic airway conductance, and measured maximal expiratory airflow at effort-independent lung volumes. CONCLUSIONS: As documented now in five cases, unsuspected microscopic mild centrilobular emphysema is the sentinel cause of loss of lung elastic recoil. This contributes significantly to expiratory airflow obstruction in never-smoking patients with asthma, with normal diffusing capacity and near-normal lung CT scan results. TRIAL REGISTRY: Protocol No. 20070934 and Study No. 1090472, Western Institutional Review Board, Olympia, WA; ClinicalTrials.gov; No. NCT00576069; URL: www.clinicaltrials.gov.

US Food and Drug Administration Approval of Whole Slide Imaging for Primary Diagnosis: A Key Milestone Is Reached and New Questions Are Raised.

April 12, 2017 marked a significant day in the evolution of digital pathology in the United States, when the US Food and Drug Administration announced its approval of the Philips IntelliSite Pathology Solution for primary diagnosis in surgical pathology. Although this event is expected to facilitate more widespread adoption of whole slide imaging for clinical applications in the United States, it also raises a number of questions as to the means by which pathologists might choose to incorporate this technology into their clinical practice. This article from the College of American Pathologists Digital Pathology Committee reviews frequently asked questions on this topic and provides answers based on currently available information.

Cocktails of human anti-cancer antibodies show a synergistic effect in nude mouse tumor xenografts.

A panel of four natural human monoclonal IgG antibodies derived from B lymphocytes isolated from regional draining lymph nodes of cancer patients has been developed and characterized. The four human antibodies are termed, RM1, RM2, RM3, and RM4. The immunoreactivity of this panel of four human antibodies is restricted to tumor cells. Individually, these human MAbs show tumor targeting and are effective in inhibiting tumor growth in nude mouse xenograft models. When used in combination the antibodies show an additive effect in slowing down the progression of tumors in xenograft models suggesting that cocktails of antibodies may be useful in the clinic.

Pathology Gross Photography: The Beginning of Digital Pathology.

The underutilized practice of photographing anatomic pathology specimens from surgical pathology and autopsies is an invaluable benefit to patients, clinicians, pathologists, and students. Photographic documentation of clinical specimens is essential for the effective practice of pathology. When considering what specimens to photograph, all grossly evident pathology, absent yet expected pathologic features, and gross-only specimens should be thoroughly documented. Specimen preparation prior to photography includes proper lighting and background, wiping surfaces of blood, removing material such as tubes or bandages, orienting the specimen in a logical fashion, framing the specimen to fill the screen, positioning of probes, and using the right-sized scale.

Pathology Gross Photography: The Beginning of Digital Pathology.

The underutilized practice of photographing anatomic pathology specimens from surgical pathology and autopsies is an invaluable benefit to patients, clinicians, pathologists, and students. Photographic documentation of clinical specimens is essential for the effective practice of pathology. When considering what specimens to photograph, all grossly evident pathology, absent yet expected pathologic features, and gross-only specimens should be thoroughly documented. Specimen preparation prior to photography includes proper lighting and background, wiping surfaces of blood, removing material such as tubes or bandages, orienting the specimen in a logical fashion, framing the specimen to fill the screen, positioning of probes, and using the right-sized scale.