Epigenetic memory of coronavirus infection in innate immune cells and their progenitors.

Inflammation can trigger lasting phenotypes in immune and non-immune cells. Whether and how human infections and associated inflammation can form innate immune memory in hematopoietic stem and progenitor cells (HSPC) has remained unclear. We found that circulating HSPC, enriched from peripheral blood, captured the diversity of bone marrow HSPC, enabling investigation of their epigenomic reprogramming following coronavirus disease 2019 (COVID-19). Alterations in innate immune phenotypes and epigenetic programs of HSPC persisted for months to 1 year following severe COVID-19 and were associated with distinct transcription factor (TF) activities, altered regulation of inflammatory programs, and durable increases in myelopoiesis. HSPC epigenomic alterations were conveyed, through differentiation, to progeny innate immune cells. Early activity of IL-6 contributed to these persistent phenotypes in human COVID-19 and a mouse coronavirus infection model. Epigenetic reprogramming of HSPC may underlie altered immune function following infection and be broadly relevant, especially for millions of COVID-19 survivors.

AI Chatbots in Clinical Laboratory Medicine: Foundations and Trends.

BACKGROUND: Artificial intelligence (AI) conversational agents, or chatbots, are computer programs designed to simulate human conversations using natural language processing. They offer diverse functions and applications across an expanding range of healthcare domains. However, their roles in laboratory medicine remain unclear, as their accuracy, repeatability, and ability to interpret complex laboratory data have yet to be rigorously evaluated. CONTENT: This review provides an overview of the history of chatbots, two major chatbot development approaches, and their respective advantages and limitations. We discuss the capabilities and potential applications of chatbots in healthcare, focusing on the laboratory medicine field. Recent evaluations of chatbot performance are presented, with a special emphasis on large language models such as the Chat Generative Pre-trained Transformer in response to laboratory medicine questions across different categories, such as medical knowledge, laboratory operations, regulations, and interpretation of laboratory results as related to clinical context. We analyze the causes of chatbots' limitations and suggest research directions for developing more accurate, reliable, and manageable chatbots for applications in laboratory medicine. SUMMARY: Chatbots, which are rapidly evolving AI applications, hold tremendous potential to improve medical education, provide timely responses to clinical inquiries concerning laboratory tests, assist in interpreting laboratory results, and facilitate communication among patients, physicians, and laboratorians. Nevertheless, users should be vigilant of existing chatbots' limitations, such as misinformation, inconsistencies, and lack of human-like reasoning abilities. To be effectively used in laboratory medicine, chatbots must undergo extensive training on rigorously validated medical knowledge and be thoroughly evaluated against standard clinical practice.

Generalizability of a Machine Learning Model for Improving Utilization of Parathyroid Hormone-Related Peptide Testing across Multiple Clinical Centers.

BACKGROUND: Measuring parathyroid hormone-related peptide (PTHrP) helps diagnose the humoral hypercalcemia of malignancy, but is often ordered for patients with low pretest probability, resulting in poor test utilization. Manual review of results to identify inappropriate PTHrP orders is a cumbersome process. METHODS: Using a dataset of 1330 patients from a single institute, we developed a machine learning (ML) model to predict abnormal PTHrP results. We then evaluated the performance of the model on two external datasets. Different strategies (model transporting, retraining, rebuilding, and fine-tuning) were investigated to improve model generalizability. Maximum mean discrepancy (MMD) was adopted to quantify the shift of data distributions across different datasets. RESULTS: The model achieved an area under the receiver operating characteristic curve (AUROC) of 0.936, and a specificity of 0.842 at 0.900 sensitivity in the development cohort. Directly transporting this model to two external datasets resulted in a deterioration of AUROC to 0.838 and 0.737, with the latter having a larger MMD corresponding to a greater data shift compared to the original dataset. Model rebuilding using site-specific data improved AUROC to 0.891 and 0.837 on the two sites, respectively. When external data is insufficient for retraining, a fine-tuning strategy also improved model utility. CONCLUSIONS: ML offers promise to improve PTHrP test utilization while relieving the burden of manual review. Transporting a ready-made model to external datasets may lead to performance deterioration due to data distribution shift. Model retraining or rebuilding could improve generalizability when there are enough data, and model fine-tuning may be favorable when site-specific data is limited.

Improved Utilization of Serial Testing Without Increased Admissions after Implementation of High-Sensitivity Troponin I: a Controlled Retrospective Cohort Study.

BACKGROUND: Guidelines recommend high-sensitivity cardiac troponin (hs-cTn) for diagnosis of myocardial infarction. Use of hs-cTn is increasing across the U.S., but questions remain regarding clinical and operational impact. Prior studies have had methodologic limitations and yielded conflicting results. OBJECTIVE: To evaluate the impact of transitioning from conventional cardiac troponin (cTn) to hs-cTn on test and resource utilization, operational efficiency, and patient safety. DESIGN: Retrospective cohort study in two New York City hospitals during the months before and after transition from conventional cTn to hs-cTn at Hospital 1. Hospital 2 served as a control. PARTICIPANTS: Consecutive emergency department (ED) patients with at least one cTn test resulted. INTERVENTION: Multifaceted hs-cTn intervention bundle, including a 0/2-h diagnostic algorithm for non-ST-elevation myocardial infarction, an educational bundle, enhancements to the electronic medical record, and nursing interventions to facilitate timed sample collection. MAIN MEASURES: Primary outcomes included serial cTn test utilization, probability of hospital admission, ED length of stay (LOS), and among discharged patients, probability of ED revisit within 72 h resulting in hospital admission. Multivariable regression models adjusted for age, sex, temporal trends, and interhospital differences. KEY RESULTS: The intervention was associated with increased use of serial cTn testing (adjusted risk difference: 48 percentage points, 95% CI: 45-50, P < 0.001) and ED LOS (adjusted geometric mean difference: 50 min, 95% CI: 50-51, P < 0.001). There was no significant association between the intervention and probability of admission (adjusted relative risk [aRR]: 0.99, 95% CI: 0.89-1.1, P = 0.81) or probability of ED revisit within 72 h resulting in admission (aRR: 1.1, 95% CI: 0.44-2.9, P = 0.81). CONCLUSIONS: Implementation of a hs-cTn intervention bundle was associated with an improvement in serial cTn testing, a neutral effect on probability of hospital admission, and a modest increase in ED LOS.

Elevated levels of renal function tests conferred increased risks of developing various pregnancy complications and adverse perinatal outcomes: insights from a population-based cohort study.

OBJECTIVES: Physiological changes during pregnancy can affect the results of renal function tests (RFTs). In this population-based cohort study, we aimed to establish trimester-specific reference intervals (RIs) of RFTs in singleton and twin pregnancies and systematically investigate the relationship between RFTs and adverse pregnancy outcomes. METHODS: The laboratory results of the first- and third-trimester RFTs, including blood urea nitrogen (BUN), serum uric acid (UA), creatinine (Crea) and cystatin C (Cys C), and the relevant medical records, were retrieved from 29,328 singleton and 840 twin pregnant women who underwent antenatal examinations from November 20, 2017 to January 31, 2021. The trimester-specific RIs of RFTs were estimated with both of the direct observational and the indirect Hoffmann methods. The associations between RTFs and pregnancy complications as well as perinatal outcomes were assessed by logistic regression analysis. RESULTS: Maternal RFTs showed no significant difference between the direct RIs established with healthy pregnant women and the calculated RIs derived from the Hoffmann method. In addition, elevated levels of RFTs were associated with increased risks of developing various pregnancy complications and adverse perinatal outcomes. Notably, elevated third-trimester RFTs posed strong risks of preterm birth (PTB) and fetal growth restriction (FGR). CONCLUSIONS: We established the trimester-specific RIs of RFTs in both singleton and twin pregnancies. Our risk analysis findings underscored the importance of RFTs in identifying women at high risks of developing adverse complications or outcomes during pregnancy.

TOP-Plus Is a Versatile Biosensor Platform for Monitoring SARS-CoV-2 Antibody Durability.

BACKGROUND: Low initial severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) antibody titers dropping to undetectable levels within months after infection have raised concerns about long-term immunity. Both the antibody levels and the avidity of the antibody-antigen interaction should be examined to understand the quality of the antibody response. METHODS: A testing-on-a-probe "plus" panel (TOP-Plus) was developed to include a newly developed avidity assay built into the previously described SARS-CoV-2 TOP assays that measured total antibody (TAb), surrogate neutralizing antibody (SNAb), IgM, and IgG on a versatile biosensor platform. TAb and SNAb levels were compared with avidity in previously infected individuals at 1.3 and 6.2 months after infection in paired samples from 80 patients with coronavirus disease 2019 (COVID-19). Sera from individuals vaccinated for SARS-CoV-2 were also evaluated for antibody avidity. RESULTS: The newly designed avidity assay in this TOP panel correlated well with a reference Bio-Layer Interferometry avidity assay (r = 0.88). The imprecision of the TOP avidity assay was <10%. Although TAb and neutralization activity (by SNAb) decreased between 1.3 and 6.2 months after infection, the antibody avidity increased significantly (P < 0.0001). Antibody avidity in 10 SARS-CoV-2 vaccinated individuals (median: 28 days after vaccination) was comparable to the measured antibody avidity in infected individuals (median: 26 days after infection). CONCLUSIONS: This highly precise and versatile TOP-Plus panel with the ability to measure SARS-CoV-2 TAb, SNAb, IgG, and IgM antibody levels and avidity of individual sera on one sensor can become a valuable asset in monitoring not only patients infected with SARS-CoV-2 but also the status of individuals' COVID-19 vaccination response.

Routine Laboratory Blood Tests Predict SARS-CoV-2 Infection Using Machine Learning.

BACKGROUND: Accurate diagnostic strategies to identify SARS-CoV-2 positive individuals rapidly for management of patient care and protection of health care personnel are urgently needed. The predominant diagnostic test is viral RNA detection by RT-PCR from nasopharyngeal swabs specimens, however the results are not promptly obtainable in all patient care locations. Routine laboratory testing, in contrast, is readily available with a turn-around time (TAT) usually within 1-2 hours. METHOD: We developed a machine learning model incorporating patient demographic features (age, sex, race) with 27 routine laboratory tests to predict an individual's SARS-CoV-2 infection status. Laboratory testing results obtained within 2 days before the release of SARS-CoV-2 RT-PCR result were used to train a gradient boosting decision tree (GBDT) model from 3,356 SARS-CoV-2 RT-PCR tested patients (1,402 positive and 1,954 negative) evaluated at a metropolitan hospital. RESULTS: The model achieved an area under the receiver operating characteristic curve (AUC) of 0.854 (95% CI: 0.829-0.878). Application of this model to an independent patient dataset from a separate hospital resulted in a comparable AUC (0.838), validating the generalization of its use. Moreover, our model predicted initial SARS-CoV-2 RT-PCR positivity in 66% individuals whose RT-PCR result changed from negative to positive within 2 days. CONCLUSION: This model employing routine laboratory test results offers opportunities for early and rapid identification of high-risk SARS-CoV-2 infected patients before their RT-PCR results are available. It may play an important role in assisting the identification of SARS-CoV-2 infected patients in areas where RT-PCR testing is not accessible due to financial or supply constraints.

Capacity for large language model chatbots to aid in orthopedic management, research, and patient queries.

Large language model (LLM) chatbots possess a remarkable capacity to synthesize complex information into concise, digestible summaries across a wide range of orthopedic subject matter. As LLM chatbots become widely available they will serve as a powerful, accessible resource that patients, clinicians, and researchers may reference to obtain information about orthopedic science and clinical management. Here, we examined the performance of three well-known and easily accessible chatbots-ChatGPT, Bard, and Bing AI-in responding to inquiries relating to clinical management and orthopedic concepts. Although all three chatbots were found to be capable of generating relevant responses, ChatGPT outperformed Bard and BingAI in each category due to its ability to provide accurate and complete responses to orthopedic queries. Despite their promising applications in clinical management, shortcomings observed included incomplete responses, lack of context, and outdated information. Nonetheless, the ability for these LLM chatbots to address these inquires has largely yet to be evaluated and will be critical for understanding the risks and opportunities of LLM chatbots in orthopedics.

The performance of artificial intelligence chatbot large language models to address skeletal biology and bone health queries.

Artificial intelligence (AI) chatbots utilizing large language models (LLMs) have recently garnered significant interest due to their ability to generate humanlike responses to user inquiries in an interactive dialog format. While these models are being increasingly utilized to obtain medical information by patients, scientific and medical providers, and trainees to address biomedical questions, their performance may vary from field to field. The opportunities and risks these chatbots pose to the widespread understanding of skeletal health and science are unknown. Here we assess the performance of 3 high-profile LLM chatbots, Chat Generative Pre-Trained Transformer (ChatGPT) 4.0, BingAI, and Bard, to address 30 questions in 3 categories: basic and translational skeletal biology, clinical practitioner management of skeletal disorders, and patient queries to assess the accuracy and quality of the responses. Thirty questions in each of these categories were posed, and responses were independently graded for their degree of accuracy by four reviewers. While each of the chatbots was often able to provide relevant information about skeletal disorders, the quality and relevance of these responses varied widely, and ChatGPT 4.0 had the highest overall median score in each of the categories. Each of these chatbots displayed distinct limitations that included inconsistent, incomplete, or irrelevant responses, inappropriate utilization of lay sources in a professional context, a failure to take patient demographics or clinical context into account when providing recommendations, and an inability to consistently identify areas of uncertainty in the relevant literature. Careful consideration of both the opportunities and risks of current AI chatbots is needed to formulate guidelines for best practices for their use as source of information about skeletal health and biology.

Artificial intelligence chatbots are increasingly used as a source of information in health care and research settings due to their accessibility and ability to summarize complex topics using conversational language. However, it is still unclear whether they can provide accurate information for questions related to the medicine and biology of the skeleton. Here, we tested the performance of three prominent chatbots­ChatGPT, Bard, and BingAI­by tasking them with a series of prompts based on well-established skeletal biology concepts, realistic physician­patient scenarios, and potential patient questions. Despite their similarities in function, differences in the accuracy of responses were observed across the three different chatbot services. While in some contexts, chatbots performed well, and in other cases, strong limitations were observed, including inconsistent consideration of clinical context and patient demographics, occasionally providing incorrect or out-of-date information, and citation of inappropriate sources. With careful consideration of their current weaknesses, artificial intelligence chatbots offer the potential to transform education on skeletal health and science.

Faithful AI in Medicine: A Systematic Review with Large Language Models and Beyond.

Artificial intelligence (AI), especially the most recent large language models (LLMs), holds great promise in healthcare and medicine, with applications spanning from biological scientific discovery and clinical patient care to public health policymaking. However, AI methods have the critical concern for generating factually incorrect or unfaithful information, posing potential long-term risks, ethical issues, and other serious consequences. This review aims to provide a comprehensive overview of the faithfulness problem in existing research on AI in healthcare and medicine, with a focus on the analysis of the causes of unfaithful results, evaluation metrics, and mitigation methods. We systematically reviewed the recent progress in optimizing the factuality across various generative medical AI methods, including knowledge-grounded LLMs, text-to-text generation, multimodality-to-text generation, and automatic medical fact-checking tasks. We further discussed the challenges and opportunities of ensuring the faithfulness of AI-generated information in these applications. We expect that this review will assist researchers and practitioners in understanding the faithfulness problem in AI-generated information in healthcare and medicine, as well as the recent progress and challenges in related research. Our review can also serve as a guide for researchers and practitioners who are interested in applying AI in medicine and healthcare.

Development of a Machine Learning Algorithm for Drug Screening Analysis on High-Resolution UPLC-MSE/QTOF Mass Spectrometry.

BACKGROUND: Ultra-performance liquid chromatography (UPLC)-MSE/quadrupole time-of-flight (QTOF) high-resolution mass spectrometry employs untargeted, data-independent acquisition in a dual mode that simultaneously collects precursor ions and product ions at low and ramped collision energies, respectively. However, algorithmic analysis of large-scale multivariate data of comprehensive drug screening as well as the positivity criteria of drug identification have not been systematically investigated. It is also unclear whether ion ratio (IR), the intensity ratio of a defined product ion divided by the precursor ion, is a stable parameter that can be incorporated into the MSE/QTOF data analysis algorithm. METHODS: IR of 91 drugs were experimentally determined and variation of IR was investigated across 5 concentrations measured on 3 different days. A data-driven machine learning approach was employed to develop multivariate linear regression (MLR) models incorporating mass error, retention time, number of detected fragment ions and IR, accuracy of isotope abundance, and peak response using drug-supplemented urine samples. Performance of the models was evaluated in an independent data set of unknown clinical urine samples in comparison with the results of manual analysis. RESULTS: IR of most compounds acquired by MSE/QTOF were low and concentration-dependent (i.e., IR increased at higher concentrations). We developed an MLR model with composite score outputs incorporating 7 parameters to predict positive drug identification. The model achieved a mean accuracy of 89.38% in the validation set and 87.92% agreement in the test set. CONCLUSIONS: The MLR model incorporating all contributing parameters can serve as a decision-support tool to facilitate objective drug identification using UPLC-MSE/QTOF.

Building the Model.

CONTEXT.­: Machine learning (ML) allows for the analysis of massive quantities of high-dimensional clinical laboratory data, thereby revealing complex patterns and trends. Thus, ML can potentially improve the efficiency of clinical data interpretation and the practice of laboratory medicine. However, the risks of generating biased or unrepresentative models, which can lead to misleading clinical conclusions or overestimation of the model performance, should be recognized. OBJECTIVES.­: To discuss the major components for creating ML models, including data collection, data preprocessing, model development, and model evaluation. We also highlight many of the challenges and pitfalls in developing ML models, which could result in misleading clinical impressions or inaccurate model performance, and provide suggestions and guidance on how to circumvent these challenges. DATA SOURCES.­: The references for this review were identified through searches of the PubMed database, US Food and Drug Administration white papers and guidelines, conference abstracts, and online preprints. CONCLUSIONS.­: With the growing interest in developing and implementing ML models in clinical practice, laboratorians and clinicians need to be educated in order to collect sufficiently large and high-quality data, properly report the data set characteristics, and combine data from multiple institutions with proper normalization. They will also need to assess the reasons for missing values, determine the inclusion or exclusion of outliers, and evaluate the completeness of a data set. In addition, they require the necessary knowledge to select a suitable ML model for a specific clinical question and accurately evaluate the performance of the ML model, based on objective criteria. Domain-specific knowledge is critical in the entire workflow of developing ML models.

Faithful AI in Medicine: A Systematic Review with Large Language Models and Beyond.

Objective: While artificial intelligence (AI), particularly large language models (LLMs), offers significant potential for medicine, it raises critical concerns due to the possibility of generating factually incorrect information, leading to potential long-term risks and ethical issues. This review aims to provide a comprehensive overview of the faithfulness problem in existing research on AI in healthcare and medicine, with a focus on the analysis of the causes of unfaithful results, evaluation metrics, and mitigation methods. Materials and Methods: Using PRISMA methodology, we sourced 5,061 records from five databases (PubMed, Scopus, IEEE Xplore, ACM Digital Library, Google Scholar) published between January 2018 to March 2023. We removed duplicates and screened records based on exclusion criteria. Results: With 40 leaving articles, we conducted a systematic review of recent developments aimed at optimizing and evaluating factuality across a variety of generative medical AI approaches. These include knowledge-grounded LLMs, text-to-text generation, multimodality-to-text generation, and automatic medical fact-checking tasks. Discussion: Current research investigating the factuality problem in medical AI is in its early stages. There are significant challenges related to data resources, backbone models, mitigation methods, and evaluation metrics. Promising opportunities exist for novel faithful medical AI research involving the adaptation of LLMs and prompt engineering. Conclusion: This comprehensive review highlights the need for further research to address the issues of reliability and factuality in medical AI, serving as both a reference and inspiration for future research into the safe, ethical use of AI in medicine and healthcare.

Explore the impact of abnormal coagulation test results on pregnancy complications and perinatal outcomes by establishing the trimester-specific reference intervals of singleton and twin pregnancies.

BACKGROUND: During pregnancy, complex physiological changes take place in the hemostatic system, resulting in a hypercoagulable state. With the established trimester-specific reference intervals (RIs) of the coagulation tests, we investigated the associations between disturbance of hemostasis and adverse pregnant outcomes in a population-based cohort study. METHODS: The first- and third-trimester coagulation tests results were retrieved from 29,328 singleton and 840 twin pregnant women for regular antenatal check-ups from November 30th, 2017 to January 31st, 2021. The trimester-specific RIs for fibrinogen (FIB), prothrombin time (PT), activated partial thromboplastin time (APTT), thrombin time (TT), d-dimer (DD) were estimated using both the direct observational and the indirect Hoffmann methods. The associations between the coagulation tests and the risks of developing pregnancy complications as well as adverse perinatal outcomes were assessed using the logistic regression analysis. RESULTS: Increased FIB, DD and decreased PT, APTT and TT were observed as the gestational age increases in the singleton pregnancy. An enhanced procoagulant state, marked by significant elevation of FIB, DD and reduction of PT, APTT and TT, was observed in the twin pregnancy. The subjects with anormal PT, APTT, TT, DD, tend to have increased risks of developing peri- and postpartum complications such as preterm birth, fetal growth restriction. CONCLUSIONS: The incidence of adverse perinatal outcomes was remarkably associated with the maternal increased levels of FIB, PT, TT, APTT and DD in the third trimester, which may be applied in early identification of women at high risk of adverse outcomes due to coagulopathy.

Chronic superphysiologic AMH promotes premature luteinization of antral follicles in human ovarian xenografts.

Anti-Müllerian hormone (AMH) is produced by growing ovarian follicles and provides a diagnostic measure of reproductive reserve in women; however, the impact of AMH on folliculogenesis is poorly understood. We cotransplanted human ovarian cortex with control or AMH-expressing endothelial cells in immunocompromised mice and recovered antral follicles for purification and downstream single-cell RNA sequencing of granulosa and theca/stroma cell fractions. A total of 38 antral follicles were observed (19 control and 19 AMH) at long-term intervals (>10 weeks). In the context of exogenous AMH, follicles exhibited a decreased ratio of primordial to growing follicles and antral follicles of increased diameter. Transcriptomic analysis and immunolabeling revealed a marked increase in factors typically noted at more advanced stages of follicle maturation, with granulosa and theca/stroma cells also displaying molecular hallmarks of luteinization. These results suggest that superphysiologic AMH alone may contribute to ovulatory dysfunction by accelerating maturation and/or luteinization of antral-stage follicles.

Measurements of SARS-CoV-2 antibody dissociation rate constant by chaotrope-free biolayer interferometry in serum of COVID-19 convalescent patients.

Kinetics measurements of antigen-antibody binding interactions are critical to understanding the functional efficiency of SARS-CoV-2 antibodies. Previously reported chaotrope-based avidity assays that rely on artificial disruption of binding do not reflect the natural binding kinetics. This study developed a chaotrope- and label-free biolayer interferometry (BLI) assay for the real-time monitoring of receptor binding domain (RBD) binding kinetics with SARS-CoV-2 spike protein in convalescent COVID-19 patients. An improved conjugation biosensor probe coated with streptavidin-polysaccharide (SA-PS) led to a six-fold increase of signal intensities and two-fold reduction of non-specific binding (NSB) compared to streptavidin only probe. Furthermore, by utilizing a separate reference probe and biotin-human serum albumin (B-HSA) blocking process to subtracted NSB signal in serum, this BLI biosensor can measure a wide range of the dissociation rate constant (koff), which can be measured without knowledge of the specific antibody concentrations. The clinical utility of this improved BLI kinetics assay was demonstrated by analyzing the koff values in sera of 24 pediatric (≤18 years old) and 63 adult (>18 years old) COVID-19 convalescent patients. Lower koff values for SARS-CoV-2 serum antibodies binding to RBD were measured in samples from children. This rapid, easy to operate and chaotrope-free BLI assay is suitable for clinical use and can be readily adapted to characterize SARS-CoV-2 antibodies developed by COVID-19 patients and vaccines.