Graph-based clinical recommender: Predicting specialists procedure orders using graph representation learning.

OBJECTIVE: To determine whether graph neural network based models of electronic health records can predict specialty consultation care needs for endocrinology and hematology more accurately than the standard of care checklists and other conventional medical recommendation algorithms in the literature. METHODS: Demand for medical expertise far outstrips supply, with tens of millions in the US alone with deficient access to specialty care. Rather than potentially months long delays to initiate diagnostic workup and medical treatment with a specialist, referring primary care supported by an automated recommender algorithm could anticipate and directly initiate patient evaluation that would otherwise be needed at subsequent a specialist appointment. We propose a novel graph representation learning approach with a heterogeneous graph neural network to model structured electronic health records and formulate recommendation/prediction of subsequent specialist orders as a link prediction problem. RESULTS: Models are trained and assessed in two specialty care sites: endocrinology and hematology. Our experimental results show that our model achieves an 8% improvement in ROC-AUC for endocrinology (ROC-AUC = 0.88) and 5% improvement for hematology (ROC-AUC = 0.84) personalized procedure recommendations over prior medical recommender systems. These recommender algorithm approaches provide medical procedure recommendations for endocrinology referrals more effectively than manual clinical checklists (recommender: precision = 0.60, recall = 0.27, F1-score = 0.37) vs. (checklist: precision = 0.16, recall = 0.28, F1-score = 0.20), and similarly for hematology referrals (recommender: precision = 0.44, recall = 0.38, F1-score = 0.41) vs. (checklist: precision = 0.27, recall = 0.71, F1-score = 0.39). CONCLUSION: Embedding graph neural network models into clinical care can improve digital specialty consultation systems and expand the access to medical experience of prior similar cases.

Signal from the noise: A mixed graphical and quantitative process mining approach to evaluate care pathways applied to emergency stroke care.

OBJECTIVE: Mapping real-world practice patterns vs. deviations from intended guidelines and protocols is necessary to identify and improve the quality of care for emergent medical conditions like acute ischemic stroke. Most status-quo process identification relies on expert opinion or direct observation, which can be biased or limited in scalability. We propose a mixed graphical and quantitative process mining approach to Electronic Health Record (EHR) event log data as a unique opportunity not only to more easily identify practice patterns, but also to compare real-world care processes and measure their conformance or variability. MATERIALS: Data was obtained from the event log of a major EHR vendor (Epic) for Stanford Health Care Hospital patients aged 18 years and older presenting to the ED from January 1, 2010 through December 31, 2018 and receiving tPA (tissue plasminogen activator) within 4.5 h of presentation. METHODS: We developed an unsupervised process-mining algorithm to create a process map from clinical event logs. The method first identifies the most common events across the cohort. Then, all possible ordered events are recorded, and a summarized vector of nodes (events) and edges (events occurring in series) are mapped by their timing and probability. The highest probability ordered pairs are used to identify the most common path. We define measures for individual pathways conformity and average conformity across all encounters. RESULTS: Automatically generated process mining graphs, and specifically it's the most common path, mimicked our institutions recommended "code stroke" clinical pathway. The average conformity score for our cohort was 0.36 (i.e. paths had an average of 36% overlap with all possible paths), with a range from high of 0.64 and low of 0.20. DISCUSSION: This method allows for unsupervised visualization of the current state of common processes as well as their most common path, which can then be used to calculate the conformity of individual pathways through this process. These results may be used to evaluate the consistency of quality care at a given institution. It may also be extended to other common processes like sepsis or myocardial infarction care or even those which currently lack standardized clinical pathways. CONCLUSION: Our mixed graphical and quantitative process mining approach represents an essential data analysis step to improve complex care processes by automatically generating qualitative and quantitative process measures from existing event log data which can then be used to target quality improvement initiatives.

Mild Cognitive Impairment: Data-Driven Prediction, Risk Factors, and Workup.

Over 78 million people will suffer from dementia by 2030, emphasizing the need for early identification of patients with mild cognitive impairment (MCI) at risk, and personalized clinical evaluation steps to diagnose potentially reversible causes. Here, we leverage real-world electronic health records in the observational medical outcomes partnership (OMOP) data model to develop machine learning models to predict MCI up to a year in advance of recorded diagnosis. Our experimental results with logistic regression, random forest, and xgboost models trained and evaluated on more than 531K patient visits show random forest model can predict MCI onset with ROC-AUC of 68.2±0.7. We identify the clinical factors mentioned in clinician notes that are most predictive of MCI. Using similar association mining techniques, we develop a data-driven list of clinical procedures commonly ordered in the workup of MCI cases, that could be used as a basis for guidelines and clinical order set templates.

Proposal for tunable dual channel transmitter and mechano-optical switch based on photonic crystal.

In this article we propose a novel mechano-optical switch and dual channel transmitter based on photonic crystal. The device consists of two waveguides and an elliptical cavity in a square lattice structure. Two optical signals at separate wavelengths are inserted in the input waveguide. The elliptical cavity can be rotated using a mechanical force, which results in the control of transmission efficiency at each of the wavelengths. In addition, rotation of the cavity can be considered as a switching action, which changes on-off states of the output signals.

Clinical Recommender Algorithms to Simulate Digital Specialty Consultations.

Advances in medical science simultaneously benefit patients while contributing to an over-whelming complexity of medicine with a decision space of thousands of possible diagnoses, tests, and treatment options. Medical expertise becomes the most important scarce health-care resource, reflected in tens of millions in the US alone with deficient access to specialty care. Combining the growing wealth of electronic medical record data with modern recommender algorithms has the potential to synthesize the clinical community's expertise into an executable format to manage this information overload and improve access to personalized care suggestions. We focus here specifically on outpatient consultations for (Endocrine) specialty expertise, one of the highest demand and most amenable areas for electronic consultation systems. Specifically we develop and evaluate models to predict the clinical orders of these initial specialty referral consultations using an ensemble of feed-forward neural networks as compared to multiple baseline algorithms. As benchmarks closer to the existing standard of care, we used diagnosis-based clinical checklists based on our review of literature and practice guidelines (e.g., Up-to-Date) for each common referral diagnosis as well as existing electronic consult referral guides. Results indicate that such automated algorithms trained on historical data can provide more personalized decision support with greater accuracy than existing benchmarks, with the potential to power fully digital consultation services that could consolidate utilization of scarce medical expertise, improving consistency of quality and access to care for more patients.

Personalized antibiograms for machine learning driven antibiotic selection.

Background: The Centers for Disease Control and Prevention identify antibiotic prescribing stewardship as the most important action to combat increasing antibiotic resistance. Clinicians balance broad empiric antibiotic coverage vs. precision coverage targeting only the most likely pathogens. We investigate the utility of machine learning-based clinical decision support for antibiotic prescribing stewardship. Methods: In this retrospective multi-site study, we developed machine learning models that predict antibiotic susceptibility patterns (personalized antibiograms) using electronic health record data of 8342 infections from Stanford emergency departments and 15,806 uncomplicated urinary tract infections from Massachusetts General Hospital and Brigham & Women's Hospital in Boston. We assessed the trade-off between broad-spectrum and precise antibiotic prescribing using linear programming. Results: We find in Stanford data that personalized antibiograms reallocate clinician antibiotic selections with a coverage rate (fraction of infections covered by treatment) of 85.9%; similar to clinician performance (84.3% p = 0.11). In the Boston dataset, the personalized antibiograms coverage rate is 90.4%; a significant improvement over clinicians (88.1% p < 0.0001). Personalized antibiograms achieve similar coverage to the clinician benchmark with narrower antibiotics. With Stanford data, personalized antibiograms maintain clinician coverage rates while narrowing 69% of empiric vancomycin+piperacillin/tazobactam prescriptions to piperacillin/tazobactam. In the Boston dataset, personalized antibiograms maintain clinician coverage rates while narrowing 48% of ciprofloxacin to trimethoprim/sulfamethoxazole. Conclusions: Precision empiric antibiotic prescribing with personalized antibiograms could improve patient safety and antibiotic stewardship by reducing unnecessary use of broad-spectrum antibiotics that breed a growing tide of resistant organisms.

Team is brain: leveraging EHR audit log data for new insights into acute care processes.

OBJECTIVE: To determine whether novel measures of contextual factors from multi-site electronic health record (EHR) audit log data can explain variation in clinical process outcomes. MATERIALS AND METHODS: We selected one widely-used process outcome: emergency department (ED)-based team time to deliver tissue plasminogen activator (tPA) to patients with acute ischemic stroke (AIS). We evaluated Epic audit log data (that tracks EHR user-interactions) for 3052 AIS patients aged 18+ who received tPA after presenting to an ED at three Northern California health systems (Stanford Health Care, UCSF Health, and Kaiser Permanente Northern California). Our primary outcome was door-to-needle time (DNT) and we assessed bivariate and multivariate relationships with six audit log-derived measures of treatment team busyness and prior team experience. RESULTS: Prior team experience was consistently associated with shorter DNT; teams with greater prior experience specifically on AIS cases had shorter DNT (minutes) across all sites: (Site 1: -94.73, 95% CI: -129.53 to 59.92; Site 2: -80.93, 95% CI: -130.43 to 31.43; Site 3: -42.95, 95% CI: -62.73 to 23.17). Teams with greater prior experience across all types of cases also had shorter DNT at two sites: (Site 1: -6.96, 95% CI: -14.56 to 0.65; Site 2: -19.16, 95% CI: -36.15 to 2.16; Site 3: -11.07, 95% CI: -17.39 to 4.74). Team busyness was not consistently associated with DNT across study sites. CONCLUSIONS: EHR audit log data offers a novel, scalable approach to measure key contextual factors relevant to clinical process outcomes across multiple sites. Audit log-based measures of team experience were associated with better process outcomes for AIS care, suggesting opportunities to study underlying mechanisms and improve care through deliberate training, team-building, and scheduling to maximize team experience.

A data value metric for quantifying information content and utility.

Data-driven innovation is propelled by recent scientific advances, rapid technological progress, substantial reductions of manufacturing costs, and significant demands for effective decision support systems. This has led to efforts to collect massive amounts of heterogeneous and multisource data, however, not all data is of equal quality or equally informative. Previous methods to capture and quantify the utility of data include value of information (VoI), quality of information (QoI), and mutual information (MI). This manuscript introduces a new measure to quantify whether larger volumes of increasingly more complex data enhance, degrade, or alter their information content and utility with respect to specific tasks. We present a new information-theoretic measure, called Data Value Metric (DVM), that quantifies the useful information content (energy) of large and heterogeneous datasets. The DVM formulation is based on a regularized model balancing data analytical value (utility) and model complexity. DVM can be used to determine if appending, expanding, or augmenting a dataset may be beneficial in specific application domains. Subject to the choices of data analytic, inferential, or forecasting techniques employed to interrogate the data, DVM quantifies the information boost, or degradation, associated with increasing the data size or expanding the richness of its features. DVM is defined as a mixture of a fidelity and a regularization terms. The fidelity captures the usefulness of the sample data specifically in the context of the inferential task. The regularization term represents the computational complexity of the corresponding inferential method. Inspired by the concept of information bottleneck in deep learning, the fidelity term depends on the performance of the corresponding supervised or unsupervised model. We tested the DVM method for several alternative supervised and unsupervised regression, classification, clustering, and dimensionality reduction tasks. Both real and simulated datasets with weak and strong signal information are used in the experimental validation. Our findings suggest that DVM captures effectively the balance between analytical-value and algorithmic-complexity. Changes in the DVM expose the tradeoffs between algorithmic complexity and data analytical value in terms of the sample-size and the feature-richness of a dataset. DVM values may be used to determine the size and characteristics of the data to optimize the relative utility of various supervised or unsupervised algorithms.

Machine Learning Predictability of Clinical Next Generation Sequencing for Hematologic Malignancies to Guide High-Value Precision Medicine.

Advancing diagnostic testing capabilities such as clinical next generation sequencing methods offer the potential to diagnose, risk stratify, and guide specialized treatment, but must be balanced against the escalating costs of healthcare to identify patient cases most likely to benefit from them. Heme-STAMP (Stanford Actionable Mutation Panel for Hematopoietic and Lymphoid Malignancies) is one such next generation sequencing test. Our objective is to assess how well Heme-STAMP pathological variants can be predicted given electronic health records data available at the time of test ordering. The model demonstrated AUROC 0.74 (95% CI: [0.72, 0.76]) with 99% negative predictive value at 6% specificity. A benchmark for comparison is the prevalence of positive results in the dataset at 58.7%. Identifying patients with very low or very high predicted probabilities of finding actionable mutations (positive result) could guide more precise high-value selection of patient cases to test.

Context is Key: Using the Audit Log to Capture Contextual Factors Affecting Stroke Care Processes.

High quality patient care through timely, precise and efficacious management depends not only on the clinical presentation of a patient, but the context of the care environment to which they present. Understanding and improving factors that affect streamlined workflow, such as provider or department busyness or experience, are essential to improving these care processes, but have been difficult to measure with traditional approaches and clinical data sources. In this exploratory data analysis, we aim to determine whether such contextual factors can be captured for important clinical processes by taking advantage of non-traditional data sources like EHR audit logs which passively track the electronic behavior of clinical teams. Our results illustrate the potential of defining multiple measures of contextual factors and their correlation with key care processes. We illustrate this using thrombolytic (tPA) treatment for ischemic stroke as an example process, but the measurement approaches can be generalized to multiple scenarios.