Implementation of a High-Value, Evidence-Based Care Program: Impact and Opportunities for Learning Organizations.

GOAL: Value-based care is not simply a matter of cost, but also one of outcomes and harms per dollar spent. This definition encompasses three key components: healthcare delivery that is organized around patients' medical conditions, costs and outcomes that are actively and consistently measured, and information technology that enables the other two components. Our objective in this project was to implement and measure a systemwide high-value, evidence-based care initiative with five pillars of high-value practices. METHODS: We performed a quasi-experimental study from September 1, 2019, to August 31, 2022, of a new care program at the University of Texas Medical Branch. Drawing from the ABIM Foundation's Choosing Wisely Campaign, the program was based on five pillars-blood management and antimicrobial, laboratory, imaging, and opioid stewardship-with interdisciplinary teams led by institutional subject matter experts (i.e., administrative leaders) accompanied by nursing, information technology, pharmacy, and clinical and nonclinical personnel including faculty and trainees. Each pillar addressed two goals with targeted interventions to assess improvements during the first three fiscal years (FYs) of implementation. The targets were set at 10% improvement by the end of each FY. Monthly measurements were recorded for each FY. PRINCIPAL FINDINGS: We tracked performance toward 30 pillar goals and determined that the teams were successful in 50%, 50%, and 70% of their goals for FY 2020, 2021, and 2022, respectively. For example, in the antimicrobial stewardship FY 2021 pillar, one goal was to decrease meropenem days of therapy (DOT) by 10% (baseline was 45 DOT/1,000 patient days; the target was 40.5 DOT/1,000 patient days). We measured quarterly DOT/1,000 patient day rates of 32.02, 30.57, and 26.9, respectively, for a cumulative rate of 26.9. Critical interventions included engaging and empowering providers and service lines (including outliers whose performance was outside norms), educational conferences, and transparent data analyses. PRACTICAL APPLICATIONS: We showed that a multidisciplinary approach to the implementation of an evidence-based, high-value care program through a partnership of engaged administrative leaders, providers, and trainees can result in sustainable and measurable high-value healthcare delivery. Specifically, structuring the program with pillars to address defined metrics resulted in progressive improvement in meeting value-based goals at the University of Texas Medical Branch. Also, challenges can be embraced as learning opportunities to inform value-based interventions that range from technological to educational tactics. The results at the University of Texas Medical Branch provide a benchmark for the implementation of a program that engages, empowers, and aligns innovative value-based care initiatives.

An interpretable multi-task system for clinically applicable COVID-19 diagnosis using CXR.

BACKGROUND: With the emergence of continuously mutating variants of coronavirus, it is urgent to develop a deep learning model for automatic COVID-19 diagnosis at early stages from chest X-ray images. Since laboratory testing is time-consuming and requires trained laboratory personal, diagnosis using chest X-ray (CXR) is a befitting option. OBJECTIVE: In this study, we proposed an interpretable multi-task system for automatic lung detection and COVID-19 screening in chest X-rays to find an alternate method of testing which are reliable, fast and easily accessible, and able to generate interpretable predictions that are strongly correlated with radiological findings. METHODS: The proposed system consists of image preprocessing and an unsupervised machine learning (UML) algorithm for lung region detection, as well as a truncated CNN model based on deep transfer learning (DTL) to classify chest X-rays into three classes of COVID-19, pneumonia, and normal. The Grad-CAM technique was applied to create class-specific heatmap images in order to establish trust in the medical AI system. RESULTS: Experiments were performed with 15,884 frontal CXR images to show that the proposed system achieves an accuracy of 91.94% in a test dataset with 2,680 images including a sensitivity of 94.48% on COVID-19 cases, a specificity of 88.46% on normal cases, and a precision of 88.01% on pneumonia cases. Our system also produced state-of-the-art outcomes with a sensitivity of 97.40% on public test data and 88.23% on a previously unseen clinical data (1,000 cases) for binary classification of COVID-19-positive and COVID-19-negative films. CONCLUSION: Our automatic computerized evaluation for grading lung infections exhibited sensitivity comparable to that of radiologist interpretation in clinical applicability. Therefore, the proposed solution can be used as one element of patient evaluation along with gold-standard clinical and laboratory testing.

Development of a prognostic model for mortality in COVID-19 infection using machine learning.

Coronavirus disease 2019 (COVID-19) is a novel disease resulting from infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), which has quickly risen since the beginning of 2020 to become a global pandemic. As a result of the rapid growth of COVID-19, hospitals are tasked with managing an increasing volume of these cases with neither a known effective therapy, an existing vaccine, nor well-established guidelines for clinical management. The need for actionable knowledge amidst the COVID-19 pandemic is dire and yet, given the urgency of this illness and the speed with which the healthcare workforce must devise useful policies for its management, there is insufficient time to await the conclusions of detailed, controlled, prospective clinical research. Thus, we present a retrospective study evaluating laboratory data and mortality from patients with positive RT-PCR assay results for SARS-CoV-2. The objective of this study is to identify prognostic serum biomarkers in patients at greatest risk of mortality. To this end, we develop a machine learning model using five serum chemistry laboratory parameters (c-reactive protein, blood urea nitrogen, serum calcium, serum albumin, and lactic acid) from 398 patients (43 expired and 355 non-expired) for the prediction of death up to 48 h prior to patient expiration. The resulting support vector machine model achieved 91% sensitivity and 91% specificity (AUC 0.93) for predicting patient expiration status on held-out testing data. Finally, we examine the impact of each feature and feature combination in light of different model predictions, highlighting important patterns of laboratory values that impact outcomes in SARS-CoV-2 infection.

Evaluating the effects of oseltamivir phosphate on platelet counts: a retrospective review.

Desialylation of platelets results in platelet clearance by the Ashwell-Morrell Receptors (AMR) found on hepatocytes. Studies suggest that oseltamivir phosphate inhibits human sialidases, enzymes responsible for desialylation, extending the lifespan of circulating platelets. We thus evaluated, the effects of oseltamivir on platelet count (PC) following treatment. Of the 385 patients evaluated for influenza, 283 (73.5%) were influenza-infected. Of the 283 infected patients, 241 (85.2%) received oseltamivir (I + O+) while 42 patients did not (I + O-). One hundred two non-infected patients received oseltamivir (I-O+). The two groups receiving oseltamivir (I + O+, I-O+), demonstrated a statistically greater increase in the PC (57.53 ± 93.81, p = .013 and 50.79 ± 70.59, p = .023, respectively) relative to the group that did not (18.45 ± 89.33 × 109/L). The observed increase in PC was statistically similar (p = .61) in both groups receiving oseltamivir (I + O+, I-O+), suggesting that this effect is independent of influenza. Comparing clinical characteristics between responders and non-responders to oseltamivir treatment showed that only duration of oseltamivir treatment (AOR = 1.30, 95% CI 1.05-1.61, p = .015) was associated with a positive PC response. Our findings suggest a correlation between oseltamivir treatment and an increase in PCs. Future studies assessing the possible uses of oseltamivir in medical conditions characterized by diminished or defective thrombopoiesis are warranted.

Multi-modal learning for inpatient length of stay prediction.

Predicting inpatient length of stay (LoS) is important for hospitals aiming to improve service efficiency and enhance management capabilities. Patient medical records are strongly associated with LoS. However, due to diverse modalities, heterogeneity, and complexity of data, it becomes challenging to effectively leverage these heterogeneous data to put forth a predictive model that can accurately predict LoS. To address the challenge, this study aims to establish a novel data-fusion model, termed as DF-Mdl, to integrate heterogeneous clinical data for predicting the LoS of inpatients between hospital discharge and admission. Multi-modal data such as demographic data, clinical notes, laboratory test results, and medical images are utilized in our proposed methodology with individual "basic" sub-models separately applied to each different data modality. Specifically, a convolutional neural network (CNN) model, which we termed CRXMDL, is designed for chest X-ray (CXR) image data, two long short-term memory networks are used to extract features from long text data, and a novel attention-embedded 1D convolutional neural network is developed to extract useful information from numerical data. Finally, these basic models are integrated to form a new data-fusion model (DF-Mdl) for inpatient LoS prediction. The proposed method attains the best R2 and EVAR values of 0.6039 and 0.6042 among competitors for the LoS prediction on the Medical Information Mart for Intensive Care (MIMIC)-IV test dataset. Empirical evidence suggests better performance compared with other state-of-the-art (SOTA) methods, which demonstrates the effectiveness and feasibility of the proposed approach.

Machine-Learning-Enabled Diagnostics with Improved Visualization of Disease Lesions in Chest X-ray Images.

The class activation map (CAM) represents the neural-network-derived region of interest, which can help clarify the mechanism of the convolutional neural network's determination of any class of interest. In medical imaging, it can help medical practitioners diagnose diseases like COVID-19 or pneumonia by highlighting the suspicious regions in Computational Tomography (CT) or chest X-ray (CXR) film. Many contemporary deep learning techniques only focus on COVID-19 classification tasks using CXRs, while few attempt to make it explainable with a saliency map. To fill this research gap, we first propose a VGG-16-architecture-based deep learning approach in combination with image enhancement, segmentation-based region of interest (ROI) cropping, and data augmentation steps to enhance classification accuracy. Later, a multi-layer Gradient CAM (ML-Grad-CAM) algorithm is integrated to generate a class-specific saliency map for improved visualization in CXR images. We also define and calculate a Severity Assessment Index (SAI) from the saliency map to quantitatively measure infection severity. The trained model achieved an accuracy score of 96.44% for the three-class CXR classification task, i.e., COVID-19, pneumonia, and normal (healthy patients), outperforming many existing techniques in the literature. The saliency maps generated from the proposed ML-GRAD-CAM algorithm are compared with the original Gran-CAM algorithm.

Evaluating Use of Generative Artificial Intelligence in Clinical Pathology Practice: Opportunities and the Way Forward.

CONTEXT.­: Generative artificial intelligence (GAI) technologies are likely to dramatically impact health care workflows in clinical pathology (CP). Applications in CP include education, data mining, decision support, result summaries, and patient trend assessments. OBJECTIVE.­: To review use cases of GAI in CP, with a particular focus on large language models. Specific examples are provided for the applications of GAI in the subspecialties of clinical chemistry, microbiology, hematopathology, and molecular diagnostics. Additionally, the review addresses potential pitfalls of GAI paradigms. DATA SOURCES.­: Current literature on GAI in health care was reviewed broadly. The use case scenarios for each CP subspecialty review common data sources generated in each subspecialty. The potential for utilization of CP data in the GAI context was subsequently assessed, focusing on issues such as future reporting paradigms, impact on quality metrics, and potential for translational research activities. CONCLUSIONS.­: GAI is a powerful tool with the potential to revolutionize health care for patients and practitioners alike. However, GAI must be implemented with much caution considering various shortcomings of the technology such as biases, hallucinations, practical challenges of implementing GAI in existing CP workflows, and end-user acceptance. Human-in-the-loop models of GAI implementation have the potential to revolutionize CP by delivering deeper, meaningful insights into patient outcomes both at an individual and population level.

How many mislabelled samples go unidentified? Results of a pilot study to determine the occult mislabelled sample rate.

BACKGROUND: Specimens with incorrect patient information are both a critical safety error and difficult to identify. Estimates of sample mislabelling rely on subjective identification of mislabelling, with the possibility that not all mislabelled samples are being caught. METHODS: We determined the blood type of two or more complete blood count specimens with the same patient label and assessed for discrepancies. We additionally determined the rate of identified sample mislabelling for the study period. RESULTS: We found a rate of 3.17 per 1000 discrepancies over the study period. These discrepancies most likely represent occult, or unidentified, mislabelled samples. In contrast, the rate of identified sample mislabelling was 1.15 per 1000. CONCLUSIONS: This study suggests that specimens identified as, or known to be, mislabelled represent only a fraction of those mislabelled. These findings are currently being confirmed in our laboratory and are likely generalisable to other institutions.

The gut microbiome-Does stool represent right?

Many stool-based gut microbiome studies have highlighted the importance of the microbiome. However, we hypothesized that stool is a poor proxy for the inner-colonic microbiome and that studying stool samples may be inadequate to capture the true inner-colonic microbiome. To test this hypothesis, we conducted prospective clinical studies with up to 20 patients undergoing an FDA-cleared gravity-fed colonic lavage without oral purgative pre-consumption. The objective of this study was to present the analysis of inner-colonic microbiota obtained non-invasively during the lavage and how these results differ from stool samples. The inner-colonic samples represented the descending, transverse, and ascending colon. All samples were analyzed for 16S rRNA and shotgun metagenomic sequences. The taxonomic, phylogenetic, and biosynthetic gene cluster analyses showed a distinctive biogeographic gradient and revealed differences between the sample types, especially in the proximal colon. The high percentage of unique information found only in the inner-colonic effluent highlights the importance of these samples and likewise the importance of collecting them using a method that can preserve these distinctive signatures. We proposed that these samples are imperative for developing future biomarkers, targeted therapeutics, and personalized medicine.

An Integrated Research-Clinical BSL-2 Platform for a Live SARS-CoV-2 Neutralization Assay.

A reliable and efficient serological test is crucial for monitoring neutralizing antibodies against SARS-CoV-2 and its variants of concern (VOCs). Here, we present an integrated research-clinical platform for a live SARS-CoV-2 neutralization assay, utilizing highly attenuated SARS-CoV-2 (Δ3678_WA1-spike). This strain contains mutations in viral transcription regulation sequences and deletion in the open-reading-frames 3, 6, 7, and 8, allowing for safe handling in biosafety level 2 (BSL-2) laboratories. Building on this backbone, we constructed a genetically stable reporter virus (mGFP Δ3678_WA1-spike) by incorporating a modified green fluorescent protein sequence (mGFP). We also constructed mGFP Δ3678_BA.5-spike and mGFP Δ3678_XBB.1.5-spike by substituting the WA1 spike with variants BA.5 and XBB.1.5 spike, respectively. All three viruses exhibit robust fluorescent signals in infected cells and neutralization titers in an optimized fluorescence reduction neutralization assay that highly correlates with a conventional plaque reduction assay. Furthermore, we established that a streamlined robot-aided Bench-to-Clinics COVID-19 Neutralization Test workflow demonstrated remarkably sensitive, specific, reproducible, and accurate characteristics, allowing the assessment of neutralization titers against SARS-CoV-2 variants within 24 h after sample receiving. Overall, our innovative approach provides a valuable avenue for large-scale testing of clinical samples against SARS-CoV-2 and VOCs at BSL-2, supporting pandemic preparedness and response strategies.

Artificial Intelligence for Vaccine Design.

Often likened to "the new electricity," artificial intelligence (AI) has broad and sweeping impact in many areas. Perhaps most exciting among these are in bioinformatics as AI allows for new and increasingly powerful ways of understanding genomics, proteomics, and immunology, just to name a few areas. Also exciting is a parallel growth in high-throughput assays including sequencing which will further accelerate the development and use of AI in biomedicine. In this chapter, we will discuss artificial intelligence and deep leaning in particular, and we will review how such approaches are enhancing and even reshaping vaccine design in terms of epitope detection and optimization. Moreover, we discuss how AI is particularly valuable to the design of mRNA vaccines including in research and production. Finally, we will discuss several additional areas across trials and operations where AI will have pervasive impact on the development of vaccines going forward.

Bioinformatic Techniques for Vaccine Development: Epitope Prediction and Structural Vaccinology.

Structural vaccinology involves characterizing the interactions between an antigen and antibodies or host immune receptors. Central to this is the task of epitope prediction, which involves describing the binding affinity and interactions of a given peptide typically to the major histocompatibility complex in the case of T-cells or to the antibodies in the case of B-cells. Several computational models exist for this purpose which we will review here. Generally, epitope predictions for MHC-I and MHC-II are substantially different tasks as well as epitope prediction for continuous versus discontinuous B-cell epitopes. Overall, these models suffer from overprediction of epitopes although general themes support both the use of neural networks as well as the incorporation of more abundant and more varied experimental annotation into model training as valuable in improving predictive performance.

Time-to-event modeling for hospital length of stay prediction for COVID-19 patients.

Providing timely patient care while maintaining optimal resource utilization is one of the central operational challenges hospitals have been facing throughout the pandemic. Hospital length of stay (LOS) is an important indicator of hospital efficiency, quality of patient care, and operational resilience. Numerous researchers have developed regression or classification models to predict LOS. However, conventional models suffer from the lack of capability to make use of typically censored clinical data. We propose to use time-to-event modeling techniques, also known as survival analysis, to predict the LOS for patients based on individualized information collected from multiple sources. The performance of six proposed survival models is evaluated and compared based on clinical data from COVID-19 patients.

Functional and Taxonomic Dysbiosis of the Gut, Urine, and Semen Microbiomes in Male Infertility.

BACKGROUND: Little is known about the role of the genitourinary and gastrointestinal microbiota in the pathogenesis of male infertility. OBJECTIVE: To compare the taxonomic and functional profiles of the gut, semen, and urine microbiomes of infertile and fertile men. DESIGN, SETTING, AND PARTICIPANTS: We prospectively enrolled 25 men with primary idiopathic infertility and 12 healthy men with proven paternity, and we collected rectal swabs, semen samples, midstream urine specimens, and experimental controls. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: We performed comprehensive semen analysis, 16S rRNA sequencing for quantitative high-resolution taxonomy, and shotgun metagenomics with a median of 140 million reads per sample for functional metabolic pathway profiling. RESULTS AND LIMITATIONS: We identified a diverse semen microbiome with modest similarity to the urinary microbiome. Infertile men harbored increased seminal α-diversity and distinct ß-diversity, increased seminal Aerococcus, and decreased rectal Anaerococcus. Prevotella abundance was inversely associated with sperm concentration, and Pseudomonas was directly associated with total motile sperm count. Vasectomy appeared to alter the seminal microbiome, suggesting a testicular or epididymal contribution. Anaerobes were highly over-represented in the semen of infertile men with a varicocele, but oxidative stress and leukocytospermia were associated with only subtle differences. Metagenomics data identified significant alterations in the S-adenosyl-L-methionine cycle, which may play a multifaceted role in the pathogenesis of infertility via DNA methylation, oxidative stress, and/or polyamine synthesis. CONCLUSIONS: This pilot study represents the first comprehensive investigation into the microbiome in male infertility. These findings provide the foundation for future investigations to explore causality and identify novel microbiome-based diagnostics and therapeutics for men with this complex and emotionally devastating disease. PATIENT SUMMARY: We explored the resident populations of bacteria living in the gut, semen, and urine of infertile and fertile men. We found several important bacterial and metabolic pathway differences with the potential to aid in diagnosing and treating male infertility in the future.