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 [...]

Abstract Background Despite the pressing need for the creation of applications that facilitate the aggregation of clinical and molecular data, most current applications are proprietary and lack the necessary compliance with standards that would allow for cross-institutional data exchange. In line with its mission of accelerating research discoveries and improving patient outcomes by linking networks of researchers, physicians, and patients focused on cancer research, caBIG (cancer Biomedical Informatics Grid™) has sponsored the creation of the caTRIP (Cancer Translational Research Informatics Platform) tool, with the purpose of aggregating clinical and molecular data in a repository that is user-friendly, easily accessible, as well as compliant with regulatory requirements of privacy and security. Results caTRIP has been developed as an N-tier architecture, with three primary tiers: domain services, the distributed query engine, and the graphical user interface, primarily making use of the caGrid infrastructure to ensure compatibility with other tools currently developed by caBIG. The application interface was designed so that users can construct queries using either the Simple Interface via drop-down menus or the Advanced Interface for more sophisticated searching strategies to using drag-and-drop. Furthermore, the application addresses the security concerns of authentication, authorization, and delegation, as well as an automated honest broker service for deidentifying data. Conclusion Currently being deployed at Duke University and a few other centers, we expect that caTRIP will make a significant contribution to further the development of translational research through the facilitation of its data exchange and storage processes.

The rapid melting of glaciers and thawing of permafrost in mountainous regions have heightened the danger of rock‐ice avalanches. These avalanches pose a severe threat due to their potential to transform into water‐saturated debris flows. The catastrophic event in Chamoli, India, on 7 February 2021, illustrates the devastating consequences of such processes. Developing a model capable of predicting the dynamics and extent of these events is imperative for natural hazard science and disaster mitigation. In response, we propose a depth‐averaged rock‐ice avalanche model encompassing four distinct materials: rock, ice, snow, and water. The model integrates crucial physical processes, including frictional heating, phase changes, ground material entrainment, and air‐blast hazards. Through a system of mass and momentum balance equations extended with grain flow and internal energy equations, the model captures heat exchanges and resulting phase changes as the fragmented material flows. Focusing on identifying the primary water source in the flow and testing the model on the 2021 Chamoli event, we quantify water's influence on flow dynamics and regime transitions. However, uncertainties persist in heat transfer physics and quantifying the hydro‐meteorological state of the flow path. Our thermo‐mechanical model enables the simulation of complex avalanches and identifies key flow transitions: powder cloud formation and potential debris flow transformation. The study underscores the pivotal role of water in avalanche dynamics and the challenge of accurately quantifying water content within the flow, necessitating comprehensive ground assessments for effective disaster management. Plain Language Summary: In the context of global climate warming, the rapid melting of glaciers and thawing of permafrost lead to rock and ice mass collapses, forming rock‐ice avalanches. These can evolve into water‐saturated debris flows, increasingly endangering societies and infrastructures in hazard‐prone regions. The 2021 Chamoli rock‐ice avalanche, which transformed into a debris flow causing over 200 fatalities, exemplifies the severe consequences of such natural disasters. To address this issue, we developed a depth‐averaged model to predict the complex dynamics characteristics of rock‐ice events. The model considers various components like rock, ice, snow, and water, along with crucial processes such as frictional heating, phase changes, material entrainment, and air‐blast hazards; and captures heat changes between phases and resultant phase transitions. Using this model, we investigated the primary water source within avalanche flows and its effects on dynamics, focusing on the 2021 Chamoli avalanche. Our findings highlight the pivotal role of water in shaping the behavior of these rock‐ice events and flow transitions. Despite significant progress, challenges remain in refining our understanding of heat transfer physics and quantifying hydro‐meteorological conditions within the flow. Nevertheless, our thermo‐mechanical model represents a significant advancement in simulating and comprehensively analyzing the dynamics of rock‐ice avalanches. Key Points: New depth‐averaged model for rock‐ice avalanches, including rock, ice, snow, and water, phases changes, ground entrainment and air blastOur model identifies three key flow transitions: powder cloud formation, debris flow transformation, and hyperconcentrated flood occurrenceModel testing shows water's impact on flow transitions. Uncertainties remain in heat transfer and entrainment calling for more validation [ABSTRACT FROM AUTHOR]

Background: Molecular testing of thyroid nodules is an essential tool to help risk stratify nodules with indeterminate cytology. Although ThyroSeq testing has been around for over a decade, there is a paucity of long‐term follow‐up data on cytologically indeterminate nodules that are determined to be molecularly negative or low‐risk. The objective of this study is to assess the outcomes of nodules with indeterminate cytology (Bethesda III or IV) and negative or low‐risk ThyroSeq results. Methods: This is a single academic institution retrospective cohort study. Patients with at least one thyroid nodule sampled with fine‐needle aspiration who underwent ThyroSeq testing from 2012 to 2018 and had negative or low‐risk ThyroSeq results on a cytologically indeterminate sample (n = 159 patients, 167 nodules) were included in the study. Outcomes include the false‐negative rate and negative predictive value of each test version, as well as follow‐up length for each nodule. Results: There were 159 patients with a mean age of 58 years (7–84 years) included in this study; the majority were female (81.8%). The mean follow‐up was 4.0 years. Of 167 nodules, three were found to be malignant on resection (1.8%). The negative predictive value for the entire cohort was 98.2% and it was 89.3% for the surgical series. Conclusion: ThyroSeq testing has good negative predictive value and can help risk stratify cytologically indeterminate nodules. Routine follow‐up allows for safe monitoring of nodules for features suggestive of malignancy. ThyroSeq testing can be used to risk stratify cytologically indeterminate nodules and decrease the number of unnecessary surgeries due to its low rates of false negatives and a high negative predictive value. [ABSTRACT FROM AUTHOR]