Use of Natural Language Processing of Patient-Initiated Electronic Health Record Messages to Identify Patients With COVID-19 Infection.

Importance: Natural language processing (NLP) has the potential to enable faster treatment access by reducing clinician response time and improving electronic health record (EHR) efficiency. Objective: To develop an NLP model that can accurately classify patient-initiated EHR messages and triage COVID-19 cases to reduce clinician response time and improve access to antiviral treatment. Design, Setting, and Participants: This retrospective cohort study assessed development of a novel NLP framework to classify patient-initiated EHR messages and subsequently evaluate the model's accuracy. Included patients sent messages via the EHR patient portal from 5 Atlanta, Georgia, hospitals between March 30 and September 1, 2022. Assessment of the model's accuracy consisted of manual review of message contents to confirm the classification label by a team of physicians, nurses, and medical students, followed by retrospective propensity score-matched clinical outcomes analysis. Exposure: Prescription of antiviral treatment for COVID-19. Main Outcomes and Measures: The 2 primary outcomes were (1) physician-validated evaluation of the NLP model's message classification accuracy and (2) analysis of the model's potential clinical effect via increased patient access to treatment. The model classified messages into COVID-19-other (pertaining to COVID-19 but not reporting a positive test), COVID-19-positive (reporting a positive at-home COVID-19 test result), and non-COVID-19 (not pertaining to COVID-19). Results: Among 10 172 patients whose messages were included in analyses, the mean (SD) age was 58 (17) years; 6509 patients (64.0%) were women and 3663 (36.0%) were men. In terms of race and ethnicity, 2544 patients (25.0%) were African American or Black, 20 (0.2%) were American Indian or Alaska Native, 1508 (14.8%) were Asian, 28 (0.3%) were Native Hawaiian or other Pacific Islander, 5980 (58.8%) were White, 91 (0.9%) were more than 1 race or ethnicity, and 1 (0.01%) chose not to answer. The NLP model had high accuracy and sensitivity, with a macro F1 score of 94% and sensitivity of 85% for COVID-19-other, 96% for COVID-19-positive, and 100% for non-COVID-19 messages. Among the 3048 patient-generated messages reporting positive SARS-CoV-2 test results, 2982 (97.8%) were not documented in structured EHR data. Mean (SD) message response time for COVID-19-positive patients who received treatment (364.10 [784.47] minutes) was faster than for those who did not (490.38 [1132.14] minutes; P = .03). Likelihood of antiviral prescription was inversely correlated with message response time (odds ratio, 0.99 [95% CI, 0.98-1.00]; P = .003). Conclusions and Relevance: In this cohort study of 2982 COVID-19-positive patients, a novel NLP model classified patient-initiated EHR messages reporting positive COVID-19 test results with high sensitivity. Furthermore, when responses to patient messages occurred faster, patients were more likely to receive antiviral medical prescription within the 5-day treatment window. Although additional analysis on the effect on clinical outcomes is needed, these findings represent a possible use case for integration of NLP algorithms into clinical care.

Effect of case and control definitions on genome-wide association study (GWAS) findings.

Genome-wide association studies (GWAS) have significantly advanced our understanding of the genetic underpinnings of diseases, but case and control cohort definitions for a given disease can vary between different published studies. For example, two GWAS for the same disease using the UK Biobank data set might use different data sources (i.e., self-reported questionnaires, hospital records, etc.) or different levels of granularity (i.e., specificity of inclusion criteria) to define cases and controls. The extent to which this variability in cohort definitions impacts the end-results of a GWAS study is unclear. In this study, we systematically evaluated the effect of the data sources used for case and control definitions on GWAS findings. Using the UK Biobank, we selected three diseases-glaucoma, migraine, and iron-deficiency anemia. For each disease, we designed 13 GWAS, each using different combinations of data sources to define cases and controls, and then calculated the pairwise genetic correlations between all GWAS for each disease. We found that the data sources used to define cases for a given disease can have a significant impact on GWAS end-results, but the extent of this depends heavily on the disease in question. This suggests the need for greater scrutiny on how case cohorts are defined for GWAS.

Regularization of Deep Neural Networks for EEG Seizure Detection to Mitigate Overfitting.

Seizure detection is a major goal for simplifying the workflow of clinicians working on EEG records. Current algorithms can only detect seizures effectively for patients already presented to the classifier. These algorithms are hard to generalize outside the initial training set without proper regularization and fail to capture seizures from the larger population. We proposed a data processing pipeline for seizure detection on an intra-patient dataset from the world's largest public EEG seizure corpus. We created spatially and session invariant features by forcing our networks to rely less on exact combinations of channels and signal amplitudes, but instead to learn dependencies towards seizure detection. For comparison, the baseline results without any additional regularization on a deep learning model achieved an F1 score of 0.544. By using random rearrangements of channels on each minibatch to force the network to generalize to other combinations of channels, we increased the F1 score to 0.629. By using random rescale of the data within a small range, we further increased the F1 score to 0.651 for our best model. Additionally, we applied adversarial multi-task learning and achieved similar results. We observed that session and patient specific dependencies were causing overfitting of deep neural networks, and the most overfitting models learnt features specific only to the EEG data presented. Thus, we created networks with regularization that the deep learning did not learn patient and session-specific features. We are the first to use random rearrangement, random rescale, and adversarial multitask learning to regularize intra-patient seizure detection and have increased sensitivity to 0.86 comparing to baseline study.

Discovery of Lipidome Alterations Following Traumatic Brain Injury via High-Resolution Metabolomics.

Traumatic brain injury (TBI) can occur across wide segments of the population, presenting in a heterogeneous manner that makes diagnosis inconsistent and management challenging. Biomarkers offer the potential to objectively identify injury status, severity, and phenotype by measuring the relative concentrations of endogenous molecules in readily accessible biofluids. Through a data-driven, discovery approach, novel biomarker candidates for TBI were identified in the serum lipidome of adult male Sprague-Dawley rats in the first week following moderate controlled cortical impact (CCI). Serum samples were analyzed in positive and negative modes by ultraperformance liquid chromatography-mass spectrometry (UPLC-MS). A predictive panel for the classification of injured and uninjured sera samples, consisting of 26 dysregulated species belonging to a variety of lipid classes, was developed with a cross-validated accuracy of 85.3% using omniClassifier software to optimize feature selection. Polyunsaturated fatty acids (PUFAs) and PUFA-containing diacylglycerols were found to be upregulated in sera from injured rats, while changes in sphingolipids and other membrane phospholipids were also observed, many of which map to known secondary injury pathways. Overall, the identified biomarker panel offers viable molecular candidates representing lipids that may readily cross the blood-brain barrier (BBB) and aid in the understanding of TBI pathophysiology.

Causes of death in the United States, 1999 to 2014.

Statistical methods have been widely used in studies of public health. Although useful in clinical research and public health policy making, these methods could not find correlation among health conditions automatically, or capture the temporal evolution of causes of death correctly. To cope with two challenges above, we implement an unsupervised machine learning model, termed topic models, to investigate the mortality data of the United States. Our model successfully groups morbidities based on their correlation, and reveals the temporal evolution of these groups from 1999 to 2014, which are also validated by existing literature. This work could provide a novel view for clinical practitioners to provide more accurate healthcare service, and for public health policymakers to make better policy.

Intraoperative Spectroscopy with Ultrahigh Sensitivity for Image-Guided Surgery of Malignant Brain Tumors.

Intraoperative cancer imaging and fluorescence-guided surgery have attracted considerable interest because fluorescence signals can provide real-time guidance to assist a surgeon in differentiating cancerous and normal tissues. Recent advances have led to the clinical use of a natural fluorophore called protoporphyrin IX (PpIX) for image-guided surgical resection of high-grade brain tumors (glioblastomas). However, traditional fluorescence imaging methods have only limited detection sensitivity and identification accuracy and are unable to detect low-grade or diffuse infiltrating gliomas (DIGs). Here we report a low-cost hand-held spectroscopic device that is capable of ultrasensitive detection of protoporphyrin IX fluorescence in vivo, together with intraoperative spectroscopic data obtained from both animal xenografts and human brain tumor specimens. The results indicate that intraoperative spectroscopy is at least 3 orders of magnitude more sensitive than the current surgical microscopes, allowing ultrasensitive detection of as few as 1000 tumor cells. For detection specificity, intraoperative spectroscopy allows the differentiation of brain tumor cells from normal brain cells with a contrast signal ratio over 100. In vivo animal studies reveal that protoporphyrin IX fluorescence is strongly correlated with both MRI and histological staining, confirming that the fluorescence signals are highly specific to tumor cells. Furthermore, ex vivo spectroscopic studies of excised brain tissues demonstrate that the hand-held spectroscopic device is capable of detecting diffuse tumor margins with low fluorescence contrast that are not detectable with current systems in the operating room. These results open new opportunities for intraoperative detection and fluorescence-guided resection of microscopic and low-grade glioma brain tumors with invasive or diffusive margins.

icuARM-An ICU Clinical Decision Support System Using Association Rule Mining.

The rapid development of biomedical monitoring technologies has enabled modern intensive care units (ICUs) to gather vast amounts of multimodal measurement data about their patients. However, processing large volumes of complex data in real-time has become a big challenge. Together with ICU physicians, we have designed and developed an ICU clinical decision support system icuARM based on associate rule mining (ARM), and a publicly available research database MIMIC-II (Multi-parameter Intelligent Monitoring in Intensive Care II) that contains more than 40,000 ICU records for 30,000+patients. icuARM is constructed with multiple association rules and an easy-to-use graphical user interface (GUI) for care providers to perform real-time data and information mining in the ICU setting. To validate icuARM, we have investigated the associations between patients' conditions such as comorbidities, demographics, and medications and their ICU outcomes such as ICU length of stay. Coagulopathy surfaced as the most dangerous co-morbidity that leads to the highest possibility (54.1%) of prolonged ICU stay. In addition, women who are older than 50 years have the highest possibility (38.8%) of prolonged ICU stay. For clinical conditions treatable with multiple drugs, icuARM suggests that medication choice can be optimized based on patient-specific characteristics. Overall, icuARM can provide valuable insights for ICU physicians to tailor a patient's treatment based on his or her clinical status in real time.

Multiscale integration of -omic, imaging, and clinical data in biomedical informatics.

This paper reviews challenges and opportunities in multiscale data integration for biomedical informatics. Biomedical data can come from different biological origins, data acquisition technologies, and clinical applications. Integrating such data across multiple scales (e.g., molecular, cellular/tissue, and patient) can lead to more informed decisions for personalized, predictive, and preventive medicine. However, data heterogeneity, community standards in data acquisition, and computational complexity are big challenges for such decision making. This review describes genomic and proteomic (i.e., molecular), histopathological imaging (i.e., cellular/tissue), and clinical (i.e., patient) data; it includes case studies for single-scale (e.g., combining genomic or histopathological image data), multiscale (e.g., combining histopathological image and clinical data), and multiscale and multiplatform (e.g., the Human Protein Atlas and The Cancer Genome Atlas) data integration. Numerous opportunities exist in biomedical informatics research focusing on integration of multiscale and multiplatform data.

Integrating image analysis algorithms in a web interface for the quantification of microtubule dynamics.

We present improvements to a web interface and an integrated computational tracking algorithm for quantitative analysis of microtubule dynamics in live-cell microscopy images. Based on a previously implemented system, more new functionalities have been added to the interface. The system also integrates a computational tracking algorithm to aid the analysis. The analysis workflow of the proposed interface is made similar to the current manual analysis workflow in order to make the interface intuitive to use. We show the workflow of the computer analysis algorithm and how it is used to aid the existing analysis workflow. We also demonstrate how to re-evaluate existing data in a case study using real imaging data. Lastly, we show the added functionalities of the interface including how to share image data and analysis results.

Cardiovascular genomics: a biomarker identification pipeline.

Genomic biomarkers are essential for understanding the underlying molecular basis of human diseases such as cardiovascular disease. In this review, we describe a biomarker identification pipeline for cardiovascular disease, which includes 1) high-throughput genomic data acquisition, 2) preprocessing and normalization of data, 3) exploratory analysis, 4) feature selection, 5) classification, and 6) interpretation and validation of candidate biomarkers. We review each step in the pipeline, presenting current and widely used bioinformatics methods. Furthermore, we analyze several publicly available cardiovascular genomics datasets to illustrate the pipeline. Finally, we summarize the current challenges and opportunities for further research.

Combining two-dimensional diffusion-ordered nuclear magnetic resonance spectroscopy, imaging desorption electrospray ionization mass spectrometry, and direct analysis in real-time mass spectrometry for the integral investigation of counterfeit pharmaceuticals.

During the past decade, there has been a marked increase in the number of reported cases involving counterfeit medicines in developing and developed countries. Particularly, artesunate-based antimalarial drugs have been targeted, because of their high demand and cost. Counterfeit antimalarials can cause death and can contribute to the growing problem of drug resistance, particularly in southeast Asia. In this study, the complementarity of two-dimensional diffusion-ordered (1)H nuclear magnetic resonance spectroscopy (2D DOSY (1)H NMR) with direct analysis in real-time mass spectrometry (DART MS) and desorption electrospray ionization mass spectrometry (DESI MS) was assessed for pharmaceutical forensic purposes. Fourteen different artesunate tablets, representative of what can be purchased from informal sources in southeast Asia, were investigated with these techniques. The expected active pharmaceutical ingredient was detected in only five formulations via both nuclear magnetic resonance (NMR) and mass spectrometry (MS) methods. Common organic excipients such as sucrose, lactose, stearate, dextrin, and starch were also detected. The graphical representation of DOSY (1)H NMR results proved very useful for establishing similarities among groups of samples, enabling counterfeit drug "chemotyping". In addition to bulk- and surface-average analyses, spatially resolved information on the surface composition of counterfeit and genuine antimalarial formulations was obtained using DESI MS that was performed in the imaging mode, which enabled one to visualize the homogeneity of both genuine and counterfeit drug samples. Overall, this study suggests that 2D DOSY (1)H NMR, combined with ambient MS, comprises a powerful suite of instrumental analysis methodologies for the integral characterization of counterfeit antimalarials.

Sphingolipidomics: a valuable tool for understanding the roles of sphingolipids in biology and disease.

The sphingolipidome is the portion of the lipidome that encompasses all sphingoid bases and their derivatives. Whereas the most studied sphingoid base is sphingosine [(2S,3R,4E)-2-aminooctadecene-1,3-diol], mammals have dozens of structural variants, and hundreds of additional types have been found in other eukaryotic organisms and some bacteria and viruses. Multiplying these figures by the N-acyl-derivatives ("ceramides") and the more than 500 phospho- and glyco- headgroups places the number of discrete molecular species in the tens of thousands or higher. Structure-specific, quantitative information about a growing fraction of the sphingolipidome can now be obtained using various types of chromatography coupled with tandem mass spectrometry, and application of these methods is producing many surprises regarding sphingolipid structure, metabolism, and function. Such large data sets can be difficult to interpret, but the development of tools that display results from genomic and lipidomic studies in a pathway relational, nodal, context can make it easier for investigators to deal with this complexity.

(Glyco)sphingolipidology: an amazing challenge and opportunity for systems biology.

Sphingolipids are found in essentially all eukaryotes and in some prokaryotes and viruses, where they influence cell structure, signaling and interactions with the extracellular environment. Because of the combinatorial nature of their biosynthesis, the sphingolipidome comprises untold thousands of species that encompass bioactive backbones and complex phospho- and glycolipids. Mass spectrometry is able to analyze a growing fraction of the sphingolipidome and is beginning to provide information about localization. Use of these structure specific, quantitative methods is producing insights, and surprises, regarding sphingolipid structure, metabolism, function and disease. Dealing with such large data sets poses special challenges for systems biology, but the intrinsic and elegant interrelationships among these compounds might provide a key to dealing with the complexity of the sphingolipidome.