Showing papers by "Mitre Corporation published in 2020"

Limited-View and Sparse Photoacoustic Tomography for Neuroimaging with Deep Learning.

Abstract: Photoacoustic tomography (PAT) is a non-ionizing imaging modality capable of acquiring high contrast and resolution images of optical absorption at depths greater than traditional optical imaging techniques. Practical considerations with instrumentation and geometry limit the number of available acoustic sensors and their “view” of the imaging target, which result in image reconstruction artifacts degrading image quality. Iterative reconstruction methods can be used to reduce artifacts but are computationally expensive. In this work, we propose a novel deep learning approach termed pixel-wise deep learning (Pixel-DL) that first employs pixel-wise interpolation governed by the physics of photoacoustic wave propagation and then uses a convolution neural network to reconstruct an image. Simulated photoacoustic data from synthetic, mouse-brain, lung, and fundus vasculature phantoms were used for training and testing. Results demonstrated that Pixel-DL achieved comparable or better performance to iterative methods and consistently outperformed other CNN-based approaches for correcting artifacts. Pixel-DL is a computationally efficient approach that enables for real-time PAT rendering and improved image reconstruction quality for limited-view and sparse PAT.

Improving Cancer Data Interoperability: The Promise of the Minimal Common Oncology Data Elements (mCODE) Initiative.

Abstract: PURPOSEBecause of expanding interoperability requirements, structured patient data are increasingly available in electronic health records. Many oncology data elements (eg, staging, biomarkers, doc...

HDDL: An Extension to PDDL for Expressing Hierarchical Planning Problems

Abstract: The research in hierarchical planning has made considerable progress in the last few years. Many recent systems do not rely on hand-tailored advice anymore to find solutions, but are supposed to be domain-independent systems that come with sophisticated solving techniques. In principle, this development would make the comparison between systems easier (because the domains are not tailored to a single system anymore) and – much more important – also the integration into other systems, because the modeling process is less tedious (due to the lack of advice) and there is no (or less) commitment to a certain planning system the model is created for. However, these advantages are destroyed by the lack of a common input language and feature set supported by the different systems. In this paper, we propose an extension to PDDL, the description language used in non-hierarchical planning, to the needs of hierarchical planning systems.

Magnetic Field Fingerprinting of Integrated-Circuit Activity with a Quantum Diamond Microscope

Abstract: Current density distributions in active integrated circuits result in patterns of magnetic fields that contain structural and functional information about the integrated circuit. Magnetic fields pass through standard materials used by the semiconductor industry and provide a powerful means to fingerprint integrated-circuit activity for security and failure analysis applications. Here, we demonstrate high spatial resolution, wide field-of-view, vector magnetic field imaging of static magnetic field emanations from an integrated circuit in different active states using a quantum diamond microscope (QDM). The QDM employs a dense layer of fluorescent nitrogen-vacancy (N-$V$) quantum defects near the surface of a transparent diamond substrate placed on the integrated circuit to image magnetic fields. We show that QDM imaging achieves a resolution of approximately $10\phantom{\rule{0.1em}{0ex}}\ensuremath{\mu}\mathrm{m}$ simultaneously for all three vector magnetic field components over the $3.7\ifmmode\times\else\texttimes\fi{}3.7\phantom{\rule{0.2em}{0ex}}{\mathrm{mm}}^{2}$ field of view of the diamond. We study activity arising from spatially dependent current flow in both intact and decapsulated field-programmable gate arrays, and find that QDM images can determine preprogrammed integrated-circuit active states with high fidelity using machine learning classification methods.

Machine Learning Approach for Flight Departure Delay Prediction and Analysis

Abstract: The expected growth in air travel demand and the positive correlation with the economic factors highlight the significant contribution of the aviation community to the US economy On‐time operati

The IMPACT of Agent Transparency on Human Performance

Abstract: The primary purpose of this article is to determine the impact of a simulated agent's transparency on human performance and related variables, such as response time, workload, and trust calibration. The agent supports participants as they complete a base defense task by managing a team of heterogeneous unmanned vehicles and serves as a decision aid to the human. Three conditions of transparency are explored. In condition 1, the agent displays only the basic information (map of vehicle location and proposed routes). In condition 2, the agent displays the basic information and an explanation of its reasoning. In condition 3, the agent displays the basic information, reasoning, and uncertainties involved in the plans. Results show that participants exhibit better performance and trust calibration in the high-transparency conditions without perceiving a significant increase in workload. However, response time also increased, likely due to the additional processing time needed for conditions with more information. Overall, our findings indicate that the increased agent transparency can improve human-agent decision making and performance, but with a small cost of the efficiency (timeliness) of task completion.

To Share is Human! Advancing Evidence into Practice through a National Repository of Interoperable Clinical Decision Support.

Abstract: Background Healthcare systems devote substantial resources to the development of clinical decision support (CDS) largely independently. The process of translating evidence-based practice into useful and effective CDS may be more efficient and less duplicative if healthcare systems shared knowledge about the translation, including workflow considerations, key assumptions made during the translation process, and technical details. Objective Describe how a national repository of CDS can serve as a public resource for healthcare systems, academic researchers, and informaticists seeking to share and reuse CDS knowledge resources or “artifacts.” Methods In 2016, the Agency for Healthcare Research and Quality (AHRQ) launched CDS Connect as a public, web-based platform for authoring and sharing CDS knowledge artifacts. Researchers evaluated early use and impact of the platform by collecting user experiences of AHRQ-sponsored and community-led dissemination efforts and through quantitative/qualitative analysis of site metrics. Efforts are ongoing to quantify efficiencies gained by healthcare systems that leverage shared, interoperable CDS artifacts rather than developing similar CDS de novo and in isolation. Results Federal agencies, academic institutions, and others have contributed over 50 entries to CDS Connect for sharing and dissemination. Analysis indicates shareable CDS resources reduce team sizes and the number of tasks and time required to design, develop, and deploy CDS. However, the platform needs further optimization to address sociotechnical challenges. Benefits of sharing include inspiring others to undertake similar CDS projects, identifying external collaborators, and improving CDS artifacts as a result of feedback. Organizations are adapting content available through the platform for continued research, innovation, and local implementations. Conclusion CDS Connect has provided a functional platform where CDS developers are actively sharing their work. CDS sharing may lead to improved implementation efficiency through numerous pathways, and further research is ongoing to quantify efficiencies gained.

A Further Study on Multiperiod Health Diagnostics Methodology under a Single-Valued Neutrosophic Set

Abstract: Employing the concept and function of tangency with similarity measures and counterpart distances for reliable medical consultations has been extensively studied in the past decades and results in lots of isomorphic measures for application. We compared the majority of such isomorphic measures proposed by various researchers and classified them into (a) maximum norm and (b) one-norm categories. Moreover, we found that previous researchers used monotonic functions to transform an identity function and resulted in complicated expressions. In this study, we provide a theoretical foundation to explain the isomorphic nature of a newer measure proposed by the following research paper against its studied existing one in deriving the same pattern recognition results. Specifically, this study initially proposes two similarity measures using maximum norm, arithmetic mean, and aggregation operators and followed by a detailed discussion on their mathematical characteristics. Subsequently, a simplified version of such measures is presented for easy application. This study completely covers two previous methods to point out that the complex approaches used were unnecessary. The findings will help physicians, patients, and their family members to obtain a proper medical diagnosis during multiple examinations.

Cryogenic operation of silicon photonic modulators based on the DC Kerr effect

Abstract: Reliable operation of photonic integrated circuits at cryogenic temperatures would enable new capabilities for emerging computing platforms, such as quantum technologies and low-power cryogenic computing. The silicon-on-insulator platform is a highly promising approach to developing large-scale photonic integrated circuits due to its exceptional manufacturability, CMOS compatibility, and high component density. Fast, efficient, and low-loss modulation at cryogenic temperatures in silicon, however, remains an outstanding challenge, particularly without the addition of exotic nonlinear optical materials. In this paper, we demonstrate DC-Kerr-effect-based modulation at a temperature of 5 K at GHz speeds, in a silicon photonic device fabricated exclusively within a CMOS-compatible process. This work opens up a path for the integration of DC Kerr modulators in large-scale photonic integrated circuits for emerging cryogenic classical and quantum computing applications.

The electronic health record as a clinical trials tool: Opportunities and challenges.

Abstract: Clinical trials provide evidence essential for progress in health care, and as the complexity of medical care has increased, the demand for such data has dramatically expanded. Conducting clinical trials has also become more complicated, evolving to meet increasing challenges in delivering clinical care and meeting regulatory requirements. Despite this, the general approach to data collection remains the same, requiring that researchers submit clinical data in response to study treatment protocols, using precisely defined data structures made available in study-specific case report forms. Currently, research data management is not integrated within the patient's clinical care record, creating added burden for clinical staff and opportunities for error. During the past decade, the electronic health record has become standard across the US healthcare system and is increasingly used to collect and analyze data reporting quality metrics for clinical care delivery. Recently, electronic health record data have also been used to address clinical research questions; however, this approach has significant drawbacks due to the unstructured and incomplete nature of current electronic health record data. This report describes steps necessary to use the electronic health record as a tool for conducting high-quality clinical research.

Status Update on Data Required to Build a Learning Health System.

Abstract: Wide adoption of electronic health records (EHRs) has raised the expectation that data obtained during routine clinical care, termed "real-world" data, will be accumulated across health care systems and analyzed on a large scale to produce improvements in patient outcomes and the use of health care resources. To facilitate a learning health system, EHRs must contain clinically meaningful structured data elements that can be readily exchanged, and the data must be of adequate quality to draw valid inferences. At the present time, the majority of EHR content is unstructured and locked into proprietary systems that pose significant challenges to conducting accurate analyses of many clinical outcomes. This article details the current state of data obtained at the point of care and describes the changes necessary to use the EHR to build a learning health system.

An overview of the effects of out-of-band interference on GNSS receivers

Abstract: This paper addresses the wide range of mechanisms through which out-of-band interference can disrupt the functioning of GNSS receivers. These mechanisms include saturation and desensitization of front-end low noise amplifiers, mixers and other circuitry; reciprocal mixing effects that arise from the fact that receivers cannot generate a perfect tone to down-convert the desired signals; intermodulation products; aliasing of out-of-band emissions that remain after filtering into the receiver's passband; and the reception of in-band (to GNSS) emissions that are always present due to imperfections in the signal generation and filtering of the interfering system. These mechanisms are described in detail and mitigation approaches for each are discussed.

Three-Dimensional Cooling of an Atom-Beam Source for High-Contrast Atom Interferometry

Abstract: We present a compact, two-stage atomic beam source that produces a continuous, narrow, collimated, and high-flux beam of rubidium atoms with sub-Doppler temperatures in three dimensions, which features very low emission of near-resonance fluorescence along the atomic trajectory. The atom-beam source originates in a pushed two-dimensional magneto-optical trap ($2{\mathrm{D}}^{+}$ MOT) feeding a slightly off-axis three-dimensional moving optical-molasses stage that continuously cools and redirects the atom beam. The capture velocity of the moving optical molasses is deliberately chosen to be low, approximately $3\phantom{\rule{0.2em}{0ex}}\mathrm{m}/\mathrm{s}$, to reduce fluorescence, and the cooling light is detuned by several atomic linewidths from resonance to reduce the absorption cross section of cooling-induced fluorescence. Near-resonance light from the $2{\mathrm{D}}^{+}$ MOT and the push beam does not propagate to the output atomic trajectory due to a ${10}^{\ensuremath{\circ}}$ bend in the atomic trajectory. The atomic beam emitted from the two-stage source has a flux up to $1.6(3)\ifmmode\times\else\texttimes\fi{}{10}^{9}\phantom{\rule{0.2em}{0ex}}\mathrm{atoms}/\mathrm{s}$, with an optimized temperature of $15.0(2)\phantom{\rule{0.2em}{0ex}}\ensuremath{\mu}\mathrm{K}$. We employ continuous Raman-Ramsey interference measurements at the atom-beam output to study the sources of decoherence in the presence of continuous cooling, and demonstrate that the atom-beam source effectively preserves high fringe contrast even during cooling. This cold-atom-beam source is appropriate for use in atom interferometers and clocks, where continuous operation eliminates dead time, the slow atom-beam velocity (6--16 m/s) improves sensitivity, the narrow 3D velocity distribution improves fringe contrast, and the low reabsorption of scattered light mitigates decoherence caused by the continuous cooling process.

Application of soft computing techniques for estimating emotional states expressed in Twitter ® time series data

Abstract: Because the emotional states of selected social groups may constitute a complex phenomenon, a suitable methodology is needed to analyze Twitter® text data that can reflect social emotions. Understanding the nature of social barometer data in terms of its underlying dynamics is critical for predicting the future states or behaviors of large social groups. This study investigated the use of the supervised soft computing techniques (1) fuzzy time series (FTS), (2) artificial neural network (ANN)-based FTS, and (3) adaptive neuro-fuzzy inference systems (ANFIS) for predicting the emotional states expressed in Twitter® data. The examined dataset contained 25,952 data points reflecting more than 380,000 Twitter® messages recorded hourly. The model prediction accuracy was performed using the root-mean-square error. The ANFIS approach resulted in the most accurate prediction among the three examined soft computing approaches. The findings of the study showed that the FTS, ANN-based FTS, and ANFIS models could be used to predict the emotional states of a large social group based on historical data. Such a modeling approach can support the development of real-time social and emotional awareness for practical decision-making, as well as rapid socio-cultural assessment and training.

Outcomes of Integrase Inhibitor-based Antiretroviral Therapy in a Clinical Cohort of Treatment-experienced Children, Adolescents and Young Adults With HIV Infection.

Abstract: BACKGROUND Data on integrase strand transfer inhibitor (INSTI) use in children, adolescents and young adults with HIV are limited. We evaluated virologic and safety outcomes following INSTI initiation among treatment-experienced children, adolescents and young adults. METHODS The DC Cohort is a multicenter observational study of individuals receiving HIV care in Washington, DC. This analysis included treatment-experienced participants 0-24 years of age who initiated an INSTI during 2011-2017. Viral suppression (VS) and safety outcomes were quantified. Differences in VS by age, sex and CD4 count were assessed using Kaplan-Meier curves. RESULTS Of 141 participants (median age 20 years; 35% 500) cells/μL were less likely to achieve VS (P < 0.001). Among participants with VS at INSTI initiation, 51% sustained VS through a median of 11.0 months of follow-up; of the 49% with transient viremia, 77% later achieved VS again. There were no safety concerns associated with the use of INSTIs. CONCLUSIONS More than half of treatment-experienced children, adolescents and young adults with detectable viremia at INSTI initiation did not achieve VS, while half of those with prior VS experienced transient viremia. Further evaluation of long-term outcomes associated with INSTI use among children, adolescents and young adults is warranted.

The Use of Design Thinking in the U.S. Federal Government

Abstract: Few organizations have more power to improve lives through innovation than the U.S. government. In this article, we focus on design thinking, a promising innovation methodology that marries a human...

COVID-19 (SARS-CoV-2) Ventilator Resource Management Using a Network Optimization Model and Predictive System Demand

Abstract: The COVID-19 (SARS-CoV-2) pandemic is overwhelming global healthcare delivery systems due to the exponential spike in cases requiring specialty tests, facilities and equipment, including complex, precision devices like ventilators. In particular, the surge in critically ill patients has revealed a significant deficiency in regional availability of respiratory care ventilators. The authors offer a mathematical framework for ventilator distribution under scarcity conditions using an optimized network model and solver. The framework is interoperable with existing COVID-19 healthcare demand models and scales for different user-defined system sizes, including hospital networks, city, state, regional and national-scale prioritization. The authors’ approach improves current capabilities for medical device resource management within the existing incident command system while accounting for availability of devices, ventilation treatment time periods, disinfection and cleaning between patients, as well as shipping logistics time. The authors present a proof of concept using a high fidelity COVID-19 data set from Colorado, discusses how to scale nationally, and emphasizes the importance of applying ethical human-in-the-loop decision making when using this or similar approaches to managing medical device resources during epidemic emergencies.

A Review of Contemporary Atomic Frequency Standards.

Abstract: Atomic frequency standards are used to generate accurate and precise time and frequency, enabling many communications, synchronization, and navigation systems in modern life. GPS and other satellite navigation systems, voice and data telecommunications, and timestamping of financial transactions all rely on precise time and frequency enabled by atomic frequency standards. This review provides a snapshot and outlook of contemporary atomic frequency standards and the applications they enable. We provide a concise summary of the performance and physics of operation of current and future atomic frequency standards. Additionally, examples of emerging frequency standard technologies and prototype demonstrations are presented, with a focus on technologies expected to provide commercial or military utility within the next decade. We include a comparison of performance vs. size and power for current atomic frequency standards, and we compare early prototypes of next-generation frequency standards to current product trends. An empirical relationship between frequency standard performance and product size is developed and discussed. Finally, we provide a mapping between applications and frequency standard technologies.

Disparities in Secure Messaging Uptake Between Patients and Physicians: Longitudinal Analysis of Two National Cross-Sectional Surveys

Abstract: Background: Emails securely exchanged between patients and clinicians offer the promise of improved access to care and indirectly improved health outcomes. Yet research to date is mixed on who—among both patients and clinicians—is using secure messaging. Objective: Using data from two large nationally representative cross-sectional surveys, this study aimed to compare the prevalence of secure messaging use among patients and their access to the functionality through their physicians, and to explore the clinical practice and physician characteristics and patient sociodemographic characteristics associated with the use of secure messaging. Methods: We conducted regression analyses to identity statistical associations between self-reported secure messaging use and access, and the patient, practice, and physician characteristics from the National Health Interview Survey (NHIS) and the National Ambulatory Medical Care Survey (NAMCS). The NHIS data collected between 2013 and 2018, with approximately 150,000 adult individuals, were used to evaluate patient characteristics associated with email communication with clinicians. The NAMCS data included 7340 physicians who reported on secure messaging use between 2013 and 2016 and provided context on physician specialty, use of certified health information technology (IT), and practice size and ownership associated with secure messaging access and use. Results: By 2016, two-thirds of ambulatory care visits were conducted by a physician who reported using secure messaging, up from 40.70% in 2013. The percentage of US residents who reported sending an email to their clinician, however, only increased from 7.22% to 16.67% between 2013 and 2018. We observed a strong positive association between certified health IT use and secure messaging use (odds ratio [OR] 11.46, 95% CI 7.55-17.39). Individuals who were black, had lower levels of education, had Medicaid or other public payer insurance, or those who were uninsured had reduced odds for using email to communicate with clinicians. No differences were observed in secure messaging use based on physician specialty, but significant differences were observed by practice size (OR 0.46, 95% CI 0.35-0.60 in solo practices vs nonsolo practices) and practice ownership (P<.001 for the different categories). Conclusions: This study is the first to use two large nationally representative surveys to produce longitudinal estimates on the access and use of patient-clinician email communication in the United States. The survey findings complement each other: one provides the patient perspective of their use and the other indicates potential patient access to secure messaging based on the use of the functionality by the physicians providing treatment. This study provides nationally representative data on the characteristics of patients and physicians who have access to and are using secure messaging. This information can be used to target interventions to promote adoption and use of secure messaging.

Overbounding GNSS/INS Integration with Uncertain GNSS Gauss-Markov Error Parameters

Abstract: The integration of GNSS with Inertial Navigation Systems (INS) has the potential to achieve high levels of continuity and availability as compared to standalone GNSS and therefore to satisfy stringent navigation requirements. However, robustly accounting for time-correlated measurement errors is a challenge when designing the Kalman filter (KF) used for GNSS/INS coupling. In particular, if the error processes are not fully known, the KF estimation error covariance can be misleading, which is problematic in safety-critical applications. In this paper, we design a GNSS/INS integration scheme that guarantees upper bounds on the estimation error variance assuming that measurement errors are first-order Gauss-Markov processes with parameters only known to reside within pre-established bounds. We evaluate the filter performance and guaranteed estimation by covariance analysis for a simulated precision approach procedure.

Wilf’s conjecture in fixed multiplicity

Abstract: We give an algorithm to determine whether Wilf’s conjecture holds for all numerical semigroups with a given multiplicity m, and use it to prove Wilf’s conjecture holds whenever m ≤ 18. Our algorith...

AI-enabled wargaming in the military decision making process

Abstract: During the Course of Action (COA) Analysis stage of the Military Decision Making Process (MDMP), staff members wargame the options of both friendly and enemy forces in an action-reaction-counteraction cycle to expose and address potential issues. This is currently a manual, subjective process, so many assumptions often go untested and only a very small number of alternative COAs may be considered. The final COA that is produced might miss opportunities or overlook risks. This challenge will only be exacerbated during Multi-Domain Operations (MDO), in which larger numbers of entities are expected to coordinate across domains to achieve converged effects within compressed timelines. This paper describes a prototype wargaming software support tool that leverages Artificial Intelligence (AI) to recommend COA improvements to commanders and staff. The tool’s design accounts for operational realities including a lack of available AI training data, limited tactical computing resources, and a need for end user interaction throughout the COA Analysis process. Given initial COAs for friendly and enemy forces, the tool searches for improvements by repeatedly proposing changes to the friendly COA and running the Data Analysis and Visualization INfrastructure for C4ISR (DAVINCI) combat simulation to evaluate them. Runtime is managed by carefully restricting the search space of the AI to only consider doctrinally relevant changes to the COA. The system architecture is designed to separate the AI, the simulation, and the user interface, simplifying continued experimentation and enhancements. The design of the AIenabled wargaming tool is presented along with initial results.

Modeling COVID-19 for lifting non-pharmaceutical interventions

Abstract: As a result of the COVID-19 worldwide pandemic, the United States instituted various non-pharmaceutical interventions (NPIs) in an effort to the slow the spread of the disease. Although necessary for public safety, these NPIs can also have deleterious effects on the economy of a nation. State and federal leaders need tools that provide insight into which combination of NPIs will have the greatest impact on slowing the disease and at what point in time it is reasonably safe to start lifting these restrictions to everyday life. In the present work, we outline a modeling process that incorporates the parameters of the disease, the effects of NPIs, and the characteristics of individual communities to offer insight into when and to what degree certain NPIs should be instituted or lifted based on the progression of a given outbreak of COVID-19.

Fragmentation and inefficiencies in US equity markets: Evidence from the Dow 30.

Abstract: Using the most comprehensive source of commercially available data on the US National Market System, we analyze all quotes and trades associated with Dow 30 stocks in calendar year 2016 from the vantage point of a single and fixed frame of reference. We find that inefficiencies created in part by the fragmentation of the equity marketplace are relatively common and persist for longer than what physical constraints may suggest. Information feeds reported different prices for the same equity more than 120 million times, with almost 64 million dislocation segments featuring meaningfully longer duration and higher magnitude. During this period, roughly 22% of all trades occurred while the SIP and aggregated direct feeds were dislocated. The current market configuration resulted in a realized opportunity cost totaling over $160 million, a conservative estimate that does not take into account intra-day offsetting events.

A New Approach for Modeling Correlated Gaussian Errors using Frequency Domain Overbounding

Abstract: This paper presents a new method to overbound Kalman filter (KF) based estimate error distributions in the presence of uncertain, time-correlated noise. Each noise component is a zero-mean Gaussian random process whose autocorrelation sequence (ACS) is stationary over the filtering duration. We show that the KF covariance matrix overbounds the estimate error distribution when the noise models overbound the Fourier transform of a windowed version of the ACS. The method is evaluated using covariance analysis for an example application in GPS-based relative position estimation.

Anticipation and Situation-Assessment Skills in Soccer Under Varying Degrees of Informational Constraint.

Abstract: The authors tested the notion that expertise effects would be more noticeable when access to situational information was reduced by occluding (i.e., noncued) or freezing (i.e., cued) the environment under temporal constraints. Using an adaptation of tasks developed by Ward, Ericsson, and Williams, the participants viewed video clips of attacking soccer plays frozen or occluded at 3 temporal points and then generated and prioritized situational options and anticipated the outcome. The high-skill players anticipated the outcomes more accurately, generated fewer task-irrelevant options, and were better at prioritizing task-relevant options than their low-skill counterparts. The anticipation scores were significantly and positively correlated with the option prioritization and task-relevant options generated but not with the total options generated. Counter to the authors’ prediction, larger skill-based option-prioritization differences were observed when the play was frozen than when it was occluded. These results indicate that processing environmental information depends on temporal and contextual conditions.

Associations Between Patient Health Outcomes and Secure Message Content Exchanged Between Patients and Clinicians: Retrospective Cohort Study

Abstract: Background: The number of electronic messages securely exchanged between clinic staff and patients has risen dramatically over the last decade. A variety of studies explored whether the volume of messages sent by patients was associated with outcomes. None of these studies, however, examined whether message content itself was associated with outcomes. Because secure messaging is a significant form of communication between patients and clinic staff, it is critical to evaluate the context of the communication to best understand its impact on patient health outcomes. Objective: To examine associations between patients’ and clinicians’ message content and changes in patients’ health outcomes. Methods: We applied a taxonomy developed specifically for secure messages to 14,394 patient- and clinic staff–generated messages derived from patient-initiated message threads. Our study population included 1602 patients, 50.94% (n=816) of whom initiated message threads. We conducted linear regression analyses to determine whether message codes were associated with changes in glycemic (A1C) levels in patients with diabetes and changes in systolic (SBP) and diastolic (DBP) blood pressure in patients with hypertension. Results: Patients who initiated threads had larger declines in A1Cs (P=.01) compared to patients who did not initiate threads. Clinic nonresponse was associated with decreased SBP (β=–.30; 95% CI –0.56 to –0.04), as were staffs’ action responses (β=–30; 95% CI –0.58 to –0.02). Increased DBP, SBP, and A1C levels were associated with patient-generated appreciation and praise messages and staff encouragement with effect sizes ranging from 0.51 (A1C) to 5.80 (SBP). We found improvements in SBP associated with patients’ complaints (β=–4.03; 95% CI –7.94 to –0.12). Deferred information sharing by clinic staff was associated with increased SBP (β=1.29; 95% CI 0.4 to 2.19). Conclusions: This is the first research to find associations between message content and patients’ health outcomes. Our findings indicate mixed associations between patient message content and patient outcomes. Further research is needed to understand the implications of this work; in the meantime, health care providers should be aware that their message content may influence patient health outcomes.