Herasevich, Vitaly

ABSTRACT Introduction Expected future delays in evacuation during near-peer conflicts in remote locales are expected to require extended care including prolonged field care over hours to days. Such delays can increase potential complications, such as insufficient blood flow (shock), bloodstream infection (sepsis), internal bleeding (hemorrhage), and require more complex treatment beyond stabilization. The Trauma Triage Treatment and Training Decision Support (4TDS) system is a real-time decision support system to monitor casualty health and identify such complications. The 4TDS software prototype operates on an Android smart phone or tablet configured for use in the DoD Nett Warrior program. It includes machine learning models to evaluate trends in six vital signs streamed from a sensor placed on a casualty to identify shock probability, internal hemorrhage risk, and need for a massive transfusion. Materials and Methods The project team used a mixed methods approach to create and evaluate the system including literature review, rapid prototyping, design requirements review, agile development, an algorithm “silent test,” and usability assessments with novice to expert medics from all three services. Results Both models, shock (showing an accuracy of 0.83) and hemorrhage/massive transfusion protocol, were successfully validated using externally collected data. All usability assessment participants completed refresher training scenarios and were able to accurately assess a simulated casualty’s condition using the phone prototype. Mean responses to statements on evaluation criteria [e.g., fit with Tactical Combat Casualty Care (TCCC), ease of use, and decision confidence] fell at five or above on a 7-point scale, indicating strong support. Conclusions Participatory design ensured 4TDS and machine learning models reflect medic and clinician mental models and work processes and built support among potential users should the system transition to operational use. Validation results can support 4TDS readiness for FDA 510k clearance as a Class II medical device.

Pulmonary auscultation is essential for detecting abnormal lung sounds during physical assessments, but its reliability depends on the operator. Machine learning (ML) models offer an alternative by automatically classifying lung sounds. ML models require substantial data, and public databases aim to address this limitation. This systematic review compares characteristics, diagnostic accuracy, concerns, and data sources of existing models in the literature. Papers published from five major databases between 1990 and 2022 were assessed. Quality assessment was accomplished with a modified QUADAS-2 tool. The review encompassed 62 studies utilizing ML models and public-access databases for lung sound classification. Artificial neural networks (ANN) and support vector machines (SVM) were frequently employed in the ML classifiers. The accuracy ranged from 49.43% to 100% for discriminating abnormal sound types and 69.40% to 99.62% for disease class classification. Seventeen public databases were identified, with the ICBHI 2017 database being the most used (66%). The majority of studies exhibited a high risk of bias and concerns related to patient selection and reference standards. Summarizing, ML models can effectively classify abnormal lung sounds using publicly available data sources. Nevertheless, inconsistent reporting and methodologies pose limitations to advancing the field, and therefore, public databases should adhere to standardized recording and labeling procedures.

OBJECTIVES: To investigate whether a novel acute care multipatient viewer (AMP), created with an understanding of clinician information and process requirements, could reduce time to clinical decision-making among clinicians caring for populations of acutely ill patients compared with a widely used commercial electronic medical record (EMR). DESIGN: Single center randomized crossover study. SETTING: Quaternary care academic hospital. SUBJECTS: Attending and in-training critical care physicians, and advanced practice providers. INTERVENTIONS: AMP. MEASUREMENTS AND MAIN RESULTS: We compared ICU clinician performance in structured clinical task completion using two electronic environments—the standard commercial EMR (Epic) versus the novel AMP in addition to Epic. Twenty subjects (10 pairs of clinicians) participated in the study. During the study session, each participant completed the tasks on two ICUs (7–10 beds each) and eight individual patients. The adjusted time for assessment of the entire ICU and the adjusted total time to task completion were significantly lower using AMP versus standard commercial EMR (–6.11; 95% CI, –7.91 to –4.30 min and –5.38; 95% CI, –7.56 to –3.20 min, respectively; p < 0.001). The adjusted time for assessment of individual patients was similar using both the EMR and AMP (0.73; 95% CI, –0.09 to 1.54 min; p = 0.078). AMP was associated with a significantly lower adjusted task load (National Aeronautics and Space Administration-Task Load Index) among clinicians performing the task versus the standard EMR (22.6; 95% CI, –32.7 to –12.4 points; p < 0.001). There was no statistically significant difference in adjusted total errors when comparing the two environments (0.68; 95% CI, 0.36–1.30; p = 0.078). CONCLUSIONS: When compared with the standard EMR, AMP significantly reduced time to assessment of an entire ICU, total time to clinical task completion, and clinician task load. Additional research is needed to assess the clinicians’ performance while using AMP in the live ICU setting.

Abstract Objectives Intensive care unit (ICU) direct care nurses spend 22% of their shift completing tasks within the electronic health record (EHR). Miscommunications and inefficiencies occur, particularly during patient hand-off, placing patient safety at risk. Redesigning how direct care nurses visualize and interact with patient information during hand-off is one opportunity to improve EHR use. A web-based survey was deployed to better understand the information and visualization needs at patient hand-off to inform redesign. Methods A multicenter anonymous web-based survey of direct care ICU nurses was conducted (9–12/2021). Semi-structured interviews with stakeholders informed survey development. The primary outcome was identifying primary EHR data needs at patient hand-off for inclusion in future EHR visualization and interface development. Secondary outcomes included current use of the EHR at patient hand-off, EHR satisfaction, and visualization preferences. Frequencies, means, and medians were calculated for each data item then ranked in descending order to generate proportional quarters using SAS v9.4. Results In total, 107 direct care ICU nurses completed the survey. The majority (46%, n = 49/107) use the EHR at patient hand-off to verify exchanged verbal information. Sixty-four percent (n = 68/107) indicated that current EHR visualization was insufficient. At the start of an ICU shift, primary EHR data needs included hemodynamics (mean 4.89 ± 0.37, 98%, n = 105), continuous IV medications (4.55 ± 0.73, 93%, n = 99), laboratory results (4.60 ± 0.56, 96%, n = 103), mechanical circulatory support devices (4.62 ± 0.72, 90%, n = 97), code status (4.40 ± 0.85, 59%, n = 108), and ventilation status (4.35 + 0.79, 51%, n = 108). Secondary outcomes included mean EHR satisfaction of 65 (0–100 scale, standard deviation = ± 21) and preferred future EHR user-interfaces to be organized by organ system (53%, n = 57/107) and visualized by tasks/schedule (61%, n = 65/107). Conclusion We identified information and visualization needs of direct care ICU nurses. The study findings could serve as a baseline toward redesigning an EHR interface.

Objective: The application of computer models in continuous patient activity monitoring using video cameras is complicated by the capture of images of varying qualities due to poor lighting conditions and lower image resolutions. Insufficient literature has assessed the effects of image resolution, color depth, noise level, and low light on the inference of eye opening and closing and body landmarks from digital images. Method: This study systematically assessed the effects of varying image resolutions (from 100 × 100 pixels to 20 × 20 pixels at an interval of 10 pixels), lighting conditions (from 42 to 2 lux with an interval of 2 lux), color-depths (from 16.7 M colors to 8 M, 1 M, 512 K, 216 K, 64 K, 8 K, 1 K, 729, 512, 343, 216, 125, 64, 27, and 8 colors), and noise levels on the accuracy and model performance in eye dimension estimation and body keypoint localization using the Dlib library and OpenPose with images from the Closed Eyes in the Wild and the COCO datasets, as well as photographs of the face captured at different light intensities. Results: The model accuracy and rate of model failure remained acceptable at an image resolution of 60 × 60 pixels, a color depth of 343 colors, a light intensity of 14 lux, and a Gaussian noise level of 4% (i.e., 4% of pixels replaced by Gaussian noise). Conclusions: The Dlib and OpenPose models failed to detect eye dimensions and body keypoints only at low image resolutions, lighting conditions, and color depths. Clinical Impact: Our established baseline threshold values will be useful for future work in the application of computer vision in continuous patient monitoring.

The study’s objective was to assess facilitators and barriers of Tele-Critical Care (TCC) perceived by SCCM members. By utilizing a survey distributed to SCCM members, a cross-sectional study was developed to analyze survey results from December 2019 and July 2020. SCCM members responded to the survey (n = 15,502) with a 1.9% response rate for the first distribution and a 2.54% response rate for the second survey (n = 9985). Participants (n = 286 and n = 254) were almost equally distributed between non-users, providers, users, and potential users of TCC services. The care delivery models for TCC were similar across most participants. Some consumers of TCC services preferred algorithmic coverage and scheduled rounds, while reactive and on-demand models were less utilized. The surveys revealed that outcome-driven measures were the principal form of TCC performance evaluation. A 1:100 (provider: patients) ratio was reported to be optimal. Factors related to costs, perceived lack of need for services, and workflow challenges were described by those who terminated TCC services. Barriers to implementation revolved around lack of reimbursement and adequate training. Interpersonal communication was identified as an essential TCC provider skill. The second survey introduced after the onset pandemic demonstrated more frequent use of advanced practice providers and focus on performance measures. Priorities for effective TCC deployment include communication, knowledge, optimal operationalization, and outcomes measurement at the organizational level. The potential effect of COVID-19 during the early stages of the pandemic on survey responses was limited and focused on the need to demonstrate TCC value.

Objective Analyzing pressure injury (PI) risk factors is complex because of multiplicity of associated factors and the multidimensional nature of this injury. The main objective of this study was to identify patients at risk of developing PI. Method Prediction performances of multiple popular supervised learning were tested. Together with the typical steps of a machine learning project, steps to prevent bias were carefully conducted, in which analysis of correlation covariance, outlier removal, confounding analysis, and cross-validation were used. Result The most accurate model reached an area under receiver operating characteristic curve of 99.7%. Ten-fold cross-validation was used to ensure that the results were generalizable. Random forest and decision tree had the highest prediction accuracy rates of 98%. Similar accuracy rate was obtained on the validation cohort. Conclusions We developed a prediction model using advanced analytics to predict PI in at-risk hospitalized patients. This will help address appropriate interventions before the patients develop a PI.