Search

* 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 pre-analytical, analytical, or post-analytical 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. (Arch Pathol Lab Med. 2024;148:e335-e361; doi: 10.5858/arpa.2023-0042-CP), Institution of machine learning (ML) in the pathology clinical domain has gained momentum and is rapidly advancing. As ML-based clinical decision support (CDS) systems continue to mature, laboratories will increasingly [...]

* Context.-Artificial intelligence algorithms hold the potential to fundamentally change many aspects of society. Application of these tools, including the publicly available ChatGPT, has demonstrated impressive domain-specific knowledge in many areas, including medicine. Objectives.-To understand the level of pathology domain-specific knowledge for ChatGPT using different underlying large language models, GPT-3.5 and the updated GPT-4. Design.-An international group of pathologists (n = 15) was recruited to generate pathology-specific questions at a similar level to those that could be seen on licensing (board) examinations. The questions (n = 15) were answered by GPT-3.5, GPT-4, and a staff pathologist who recently passed their Canadian pathology licensing exams. Participants were instructed to score answers on a 5-point scale and to predict which answer was written by ChatGPT. Results.-GPT-3.5 performed at a similar level to the staff pathologist, while GPT-4 outperformed both. The overall score for both GPT-3.5 and GPT-4 was within the range of meeting expectations for a trainee writing licensing examinations. In all but one question, the reviewers were able to correctly identify the answers generated by GPT-3.5. Conclusions.-By demonstrating the ability of ChatGPT to answer pathology-specific questions at a level similar to (GPT-3.5) or exceeding (GPT-4) a trained pathologist, this study highlights the potential of large language models to be transformative in this space. In the future, more advanced iterations of these algorithms with increased domain-specific knowledge may have the potential to assist pathologists and enhance pathology resident training. (Arch Pathol Lab Med. 2024;148:1152-1158; doi: 10.5858/arpa.2023-0296-OA), Rapid technological advancements in molecular pathology and a continuously growing body of knowledge have resulted in an unprecedented increase in diagnostic complexity for pathologists. (1) This presents a challenge for [...]

* Context.--Breast carcinoma grade, as determined by the Nottingham Grading System (NGS), is an important criterion for determining prognosis. The NGS is based on 3 parameters: tubule formation (TF), nuclear pleomorphism (NP), and mitotic count (MC). The advent of digital pathology and artificial intelligence (AI) have increased interest in virtual microscopy using digital whole slide imaging (WSI) more broadly. Objective.--To compare concordance in breast carcinoma grading between AI and a multi-institutional group of breast pathologists using digital WSI. Design.--We have developed an automated NGS framework using deep learning. Six pathologists and AI independently reviewed a digitally scanned slide from 137 invasive carcinomas and assigned a grade based on scoring of the TF, NP, and MC. Results.--Interobserver agreement for the pathologists and AI for overall grade was moderate ([kappa] = 0.471). Agreement was good ([kappa] = 0.681), moderate ([kappa] = 0.442), and fair ([kappa] = 0.368) for grades 1, 3, and 2, respectively. Observer pair concordance for AI and individual pathologists ranged from fair to good ([kappa] = 0.313-0.606). Perfect agreement was observed in 25 cases (27.4%). Interobserver agreement for the individual components was best for TF ([kappa] = 0.471 each) followed by NP ([kappa] = 0.342) and was worst for MC ([kappa] = 0.233). There were no observed differences in concordance amongst pathologists alone versus pathologists + AI. Conclusions.--Ours is the first study comparing concordance in breast carcinoma grading between a multi-institutional group of pathologists using virtual microscopy to a newly developed WSI AI methodology. Using explainable methods, AI demonstrated similar concordance to pathologists alone. doi: 10.5858/arpa.2021-0299-OA, Over the years, microscopic evaluation of hematoxylin-eosin-stained slides from patient samples has been the gold standard for cancer diagnosis and grading of malignancy. The Nottingham Grading System (NGS) is widely [...]

* Context.--Studies on the adoption of voice recognition in health care have mostly focused on turnaround time and error rate, with less attention paid to the impact on the efficiency of the providers. Objective.--To study the impact of voice recognition on the efficiency of grossing biopsy specimens. Design.--Timestamps corresponding to barcode scanning for biopsy specimen bottles and cassettes were retrieved from the pathology information system database. The time elapsed between scanning a specimen bottle and the corresponding first cassette was the length of time spent on the gross processing of that specimen and is designated as the specimen time. For the first specimen of a case, the specimen time additionally included the time spent on dictating the clinical information. Therefore, the specimen times were divided into the following 2 categories: first-specimen time and subsequent-specimen time. The impact of voice recognition on specimen times was studied using both univariate and multivariate analyses. Results.--Specimen complexity, prosector variability, length of clinical information text, and the number of biopsies the prosector grossed that day were the major determinants of specimen times. Adopting voice recognition had a negligible impact on specimen times. Conclusions.--Adopting voice recognition in the gross room removes the need to hire transcriptionists without negatively impacting the efficiency of the prosectors, resulting in an overall cost saving. Using computer scripting to automatically enter clinical information (received through the electronic order interface) into report templates may potentially increase the grossing efficiency in the future. doi: 10.5858/arpa.2020-0115-OA, The adoption of voice recognition technology into the surgical pathology workflow has been occurring during the last 2 decades. (1-4) Publications on the effects of incorporating voice recognition have mostly [...]

In the last decade there has been a rapid advancement in the use of information and communication technologies to empower the practice of medicine. Wearable smart glasses, such as Google [...], Context.--Smart glasses are a wearable technology that enable hands-free data acquisition and entry. Objective.--To develop a surgical pathology grossing application on a smart glass platform. Design.--An existing logistics software for the Google Glass Enterprise smart glass platform was used to create surgical pathology grossing protocols. The 2 grossing protocols were developed to simulate grossing a complex (heart) and a simple (kidney) specimen. For both protocols, users were visually prompted by the smart glass device to perform each task, record measurements, or document the field of view. In addition to measuring the total time of the protocol performance, each substep within the protocol was automatically recorded. Subsequently, a report was generated that contained the dictation, images, voice recordings, and the timing of each step. The application was tested by 3 users using the 2 grossing protocols. The users were tracked across 3 grossing procedures for each protocol. Results.--For the complex specimen grossing the average time across repeated procedures was not significantly different between users (P > .99). However, when grossing times of the complex specimen were compared for repeated performances of the same user, a significant reduction in grossing times was observed with each repetition (P = .002). For the simple specimen, the average grossing time across multiple attempts was different among users (P = .03); however, no improvement in grossing time was observed with repeated performance (P = .499). Conclusions.--Augmented reality based grossing applications can provide automated data collection to track the changes in grossing performance over time. doi: 10.5858/arpa.2020-0090-OA

In routine clinical practice, pathologists may encounter extraneous pieces of tissue on glass slides that could be because of contamination from other specimens. These are typically called 'tissue floaters.' A [...], Context.--Pathologists may encounter extraneous pieces of tissue (tissue floaters) on glass slides because of specimen cross-contamination. Troubleshooting this problem, including performing molecular tests for tissue identification if available, is time consuming and often does not satisfactorily resolve the problem. Objective.--To demonstrate the feasibility of using an image search tool to resolve the tissue floater conundrum. Design.--A glass slide was produced containing 2 separate hematoxylin and eosin (H&E)-stained tissue floaters. This fabricated slide was digitized along with the 2 slides containing the original tumors used to create these floaters. These slides were then embedded into a dataset of 2325 whole slide images comprising a wide variety of H&E stained diagnostic entities. Digital slides were broken up into patches and the patch features converted into barcodes for indexing and easy retrieval. A deep learning-based image search tool was employed to extract features from patches via barcodes, hence enabling image matching to each tissue floater. Results.--There was a very high likelihood of finding a correct tumor match for the queried tissue floater when searching the digital database. Search results repeatedly yielded a correct match within the top 3 retrieved images. The retrieval accuracy improved when greater proportions of the floater were selected. The time to run a search was completed within several milliseconds. Conclusions.--Using an image search tool offers pathologists an additional method to rapidly resolve the tissue floater conundrum, especially for those laboratories that have transitioned to going fully digital for primary diagnosis. (Arch Pathol Lab Med. 2021;145:359-364; doi: 10.5858/arpa.2020-0034-OA)

On December 29, 2019, official reports describing a cluster of patients suffering from a severe novel pneumonia of unknown etiology initially surfaced from Wuhan (Hubei province) in the People's Republic [...], * Context.--The coronavirus disease 19 (COVID-19) pandemic is placing unparalleled burdens on regional and institutional resources in medical facilities across the globe. This disruption is causing unprecedented downstream effects to traditionally established channels of patient care delivery, including those of essential anatomic pathology services. With Washington state being the initial North American COVID-19 epicenter, the University of Washington in Seattle has been at the forefront of conceptualizing and implementing innovative solutions in order to provide uninterrupted quality patient care amidst this growing crisis. Objective.--To conduct a rapid validation study assessing our ability to reliably provide diagnostic neuropathology services via a whole slide imaging (WSI) platform as part of our departmental COVID-19 planning response. Design.--This retrospective study assessed diagnostic concordance of neuropathologic diagnoses rendered via WSI as compared to those originally established via traditional histopathology in a cohort of 30 cases encompassing a broad range of neurosurgical and neuromuscular entities. This study included the digitalization of 93 slide preparations, which were independently examined by groups of board-certified neuropathologists and neuropathology fellows. Results.--There were no major or minor diagnostic discrepancies identified in either the attending neuropathologist or neuropathology trainee groups for either the neurosurgical or neuromuscular case cohorts. Conclusions.--Our study demonstrates that accuracy of neuropathologic diagnoses and interpretation of ancillary preparations via WSI are not inferior to those generated via traditional microscopy. This study provides a framework for rapid subspecialty validation and deployment of WSI for diagnostic purposes during a pandemic event. doi: 10.5858/arpa.2020-0372-SA

Coronavirus Disease 2019 (COVID-19) emerged as one of the largest pandemics within the last century, leading to a swift upending of American society. Many employees faced a shift to working [...]

* Context.--Nuclear grooves and inclusions are major features of cancer. However, the nuclear irregularities in renal cell carcinoma (RCC) have not yet been well characterized. Objective.--To determine the clinicopathologic significance of nuclear grooves and inclusions in RCC. Design.--The frequencies or scores of nuclear grooves and inclusions were compared with the histologic subtype, nuclear grade, and TNM stage, as well as overall survival of RCC patients. For objective counting of nuclear irregularities, a relational image database was constructed and used for quantitative assessment. Results.--Nuclear grooves and inclusions were seen in 96% and 65% of 110 RCC cases, respectively. The intranuclear inclusions were found more frequently in chromophobe and papillary types than in clear cell carcinoma (P < .001). The nuclear scores, the sum of grooves or inclusions per 5000 tumor cells, were highly related to the histologic subtype (P < .001). Clear cell RCCs with high inclusion scores (2 or more) were correlated with poorer overall survival in comparison to clear cell carcinomas with low inclusion scores (P = .04). The groove scores were highly associated with Fuhrman grade (P = .003) but not with overall survival of clear cell RCC patients (P = .65). In multivariate analysis, higher inclusion scores and advanced tumor stages (III/IV) were correlated with worse outcomes of clear cell RCC. Conclusions.--Nuclear grooves and inclusions are histologic components of RCCs, especially chromophobe and papillary carcinomas. Furthermore, nuclear inclusions might be an independent prognostic factor for clear cell RCC. (Arch Pathol Lab Med. 2008;132:940-946), Nuclear grooves and inclusions are important diagnostic features of cancer, although intranuclear inclusions are seen in normal epithelial cells. (1) In fact, nuclear inclusions are commonly found in some benign [...]

To place a cloak of reality on business prognostications, the first questions to answer are (1) what is the real business in consideration, (2) what are the components or analogs [...]

I have been asked to talk about reality testing of educational aspects of new and emerging imaging technologies. Newer technologies, such as telepathology virtual slides (whole slide digital imaging), may [...]

The clinical practice of anatomic pathology and radiology are similar at several levels. Both disciplines require an imaging modality to collect primary data. For radiology, the modality can be varied [...]

The struggle of today, is not altogether for today-it is for a vast future also. Abraham Lincoln (1809-1865), US president This article is an overview of a presentation from the [...]