Applications of Artificial Intelligence in Helicopter Emergency Medical Services: A Scoping Review.

OBJECTIVE: Recent systematic reviews of acute care medicine applications of artificial intelligence (AI) have focused on hospital and general prehospital uses. The purpose of this scoping review was to identify and describe the literature on AI use with a focus on applications in helicopter emergency medical services (HEMS). METHODS: A literature search was performed with specific inclusion and exclusion criteria. Articles were grouped by characteristics such as publication year and general subject matter with categoric and temporal trend analyses. RESULTS: We identified 21 records focused on the use of AI in HEMS. These applications included both clinical and triage uses and nonclinical uses. The earliest study appeared in 2006, but over one third of the identified studies have been published in 2021 or later. The passage of time has seen an increased likelihood of HEMS AI studies focusing on nonclinical issues; for each year, the likelihood of a nonclinical focus had an odds ratio of 1.3. CONCLUSION: This scoping review provides overview and hypothesis-generating information regarding AI applications specific to HEMS. HEMS AI may be ultimately deployed in nonclinical arenas as much as or more than for clinical decision support. Future studies will inform future decisions as to how AI may improve HEMS systems design, asset deployment, and clinical care.

Hypotension During Vasopressor Infusion Occurs in Predictable Clusters: A Multicenter Analysis.

Background: Published evidence indicates that mean arterial pressure (MAP) below a goal range (hypotension) is associated with worse outcomes, though MAP management failures are common. We sought to characterize hypotension occurrences in ICUs and consider the implications for MAP management. Methods: Retrospective analysis of 3 hospitals' cohorts of adult ICU patients during continuous vasopressor infusion. Two cohorts were general, mixed ICU patients and one was exclusively acute spinal cord injury patients. "Hypotension-clusters" were defined where there were ≥10 min of cumulative hypotension over a 60-min period and "constant hypotension" was ≥10 continuous minutes. Trend analysis was performed (predicting future MAP using 14 min of preceding MAP data) to understand which hypotension-clusters could likely have been predicted by clinician awareness of MAP trends. Results: In cohorts of 155, 66, and 16 ICU stays, respectively, the majority of hypotension occurred within the hypotension-clusters. Failures to keep MAP above the hypotension threshold were notable in the bottom quartiles of each cohort, with hypotension durations of 436, 167, and 468 min, respectively, occurring within hypotension-clusters per day. Mean arterial pressure trend analysis identified most hypotension-clusters before any constant hypotension occurred (81.2%-93.6% sensitivity, range). The positive predictive value of hypotension predictions ranged from 51.4% to 72.9%. Conclusions: Across 3 cohorts, most hypotension occurred in temporal clusters of hypotension that were usually predictable from extrapolation of MAP trends.

Diagnostic suspicion bias and machine learning: Breaking the awareness deadlock for sepsis detection.

Many early warning algorithms are downstream of clinical evaluation and diagnostic testing, which means that they may not be useful when clinicians fail to suspect illness and fail to order appropriate tests. Depending on how such algorithms handle missing data, they could even indicate "low risk" simply because the testing data were never ordered. We considered predictive methodologies to identify sepsis at triage, before diagnostic tests are ordered, in a busy Emergency Department (ED). One algorithm used "bland clinical data" (data available at triage for nearly every patient). The second algorithm added three yes/no questions to be answered after the triage interview. Retrospectively, we studied adult patients from a single ED between 2014-16, separated into training (70%) and testing (30%) cohorts, and a final validation cohort of patients from four EDs between 2016-2018. Sepsis was defined per the Rhee criteria. Investigational predictors were demographics and triage vital signs (downloaded from the hospital EMR); past medical history; and the auxiliary queries (answered by chart reviewers who were blinded to all data except the triage note and initial HPI). We developed L2-regularized logistic regression models using a greedy forward feature selection. There were 1164, 499, and 784 patients in the training, testing, and validation cohorts, respectively. The bland clinical data model yielded ROC AUC's 0.78 (0.76-0.81) and 0.77 (0.73-0.81), for training and testing, respectively, and ranged from 0.74-0.79 in four hospital validation. The second model which included auxiliary queries yielded 0.84 (0.82-0.87) and 0.83 (0.79-0.86), and ranged from 0.78-0.83 in four hospital validation. The first algorithm did not require clinician input but yielded middling performance. The second showed a trend towards superior performance, though required additional user effort. These methods are alternatives to predictive algorithms downstream of clinical evaluation and diagnostic testing. For hospital early warning algorithms, consideration should be given to bias and usability of various methods.

APPRAISE-HRI: AN ARTIFICIAL INTELLIGENCE ALGORITHM FOR TRIAGE OF HEMORRHAGE CASUALTIES.

ABSTRACT: Background: Hemorrhage remains the leading cause of death on the battlefield. This study aims to assess the ability of an artificial intelligence triage algorithm to automatically analyze vital-sign data and stratify hemorrhage risk in trauma patients. Methods: Here, we developed the APPRAISE-Hemorrhage Risk Index (HRI) algorithm, which uses three routinely measured vital signs (heart rate and diastolic and systolic blood pressures) to identify trauma patients at greatest risk of hemorrhage. The algorithm preprocesses the vital signs to discard unreliable data, analyzes reliable data using an artificial intelligence-based linear regression model, and stratifies hemorrhage risk into low (HRI:I), average (HRI:II), and high (HRI:III). Results: To train and test the algorithm, we used 540 h of continuous vital-sign data collected from 1,659 trauma patients in prehospital and hospital (i.e., emergency department) settings. We defined hemorrhage cases (n = 198) as those patients who received ≥1 unit of packed red blood cells within 24 h of hospital admission and had documented hemorrhagic injuries. The APPRAISE-HRI stratification yielded a hemorrhage likelihood ratio (95% confidence interval) of 0.28 (0.13-0.43) for HRI:I, 1.00 (0.85-1.15) for HRI:II, and 5.75 (3.57-7.93) for HRI:III, suggesting that patients categorized in the low-risk (high-risk) category were at least 3-fold less (more) likely to have hemorrhage than those in the average trauma population. We obtained similar results in a cross-validation analysis. Conclusions: The APPRAISE-HRI algorithm provides a new capability to evaluate routine vital signs and alert medics to specific casualties who have the highest risk of hemorrhage, to optimize decision-making for triage, treatment, and evacuation.

Predicting Hypertensive Events with Time-Series Analysis of Mean Arterial Pressure.

We investigated whether a statistical model used previously to predict hypotension from mean arterial pressure (MAP) time series analysis could predict hypertension. We performed a retrospective analysis of minute-by-minute MAP records from two cohorts of intensive care unit (ICU) patients. The first cohort was comprised of surgical and medical ICUs while the second cohort was comprised of acute spinal cord injury (ASCI) patients in a neurological ICU. At each time point with physiological MAP, time series analysis was used to predict the median MAP for the subsequent 20 min. This method was used to predict hypertensive episodes, i.e., intervals of 20 or more minutes where at least half of the MAP measurements were > 105 mmHg. Advance prediction of hypertensive episodes was similar in the two cohorts (69.15% vs. 82.61%, respectively), as was positive predictive value of the hypertension predictions (67.42% vs. 71.57%). The results suggest that the methodology may be useable for predicting hypertension from time-series analysis of MAP. Patients requiring continuous vasopressor infusion are at risk of hypertension and excessive vasoconstriction. We found evidence that time-series analysis previously validated for predicting hypotension may also be usable for predicting hypertension.

End-user evaluation of an interface for clinical decision support using predictive algorithms.

There have been decades of interest in advanced computational algorithms with potential for clinical decision support systems (CDSS), yet these have not been widely implemented in clinical practice. One major barrier to dissemination may be a user-friendly interface that integrates into clinical workflows. Complicated or non-intuitive displays may confuse users and may even increase patient management errors. We recently developed a graphical user interface (GUI) intended to integrate a predictive hemodynamic model into the workflow of nurses caring for patients on vasopressors in the intensive care unit (ICU). Here, we evaluated user perceptions of the usability of this system. The software was installed in the room of an ICU patient, running for at least 4 hours with the display hidden. Afterward, we showed nurses a video recording of the session and surveyed their perceptions about the software's potential safety and usefulness. We collected data for nine patients. Overall, nurses expressed reasonable enthusiasm that the software would be useful and without serious safety concerns. However, there was a wide diversity of opinions about what specific aspects of the software would be useful and what aspects were confusing. In several instances, the same elements of the GUI were cited as most useful by some nurses and most confusing by others. Our findings validate that it is possible to develop GUIs for CDSS that are perceived as potentially useful and without substantial risk but also reinforce the diversity of user perceptions about novel CDSS technology. Clinical Relevance- This end-user evaluation of a novel CDSS highlights the importance of end-user experience in the workflow integration of advanced computational algorithms for bedside decision support during critical care.

Modeling of Usual Care: Vasopressor Initiation for Sepsis With Hypotension.

Usual care regarding vasopressor initiation is ill-defined. We aimed to develop a quantitative "dynamic practice" model for usual care in the emergency department (ED) regarding the timing of vasopressor initiation in sepsis. In a retrospective study of 589 septic patients with hypotension in an urban tertiary care center ED, we developed a multi-variable model that distinguishes between patients who did and did not subsequently receive sustained (>24 h) vasopressor therapy. Candidate predictors were vital signs, intravenous fluid (IVF) volumes, laboratory measurements, and elapsed time from triage computed at timepoints leading up to the final decision timepoint of either vasopressor initiation or ED hypotension resolution without vasopressors. A model with six independently significant covariates (respiratory rate, Glasgow Coma Scale score, SBP, SpO2, administered IVF, and elapsed time) achieved a C-statistic of 0.78 in a held-out test set at the final decision timepoint, demonstrating the ability to reliably model usual care for vasopressor initiation for hypotensive septic patients. The included variables measured depth of hypotension, extent of disease severity and organ dysfunction. At an operating point of 90% specificity, the model identified a minority of patients (39%) more than an hour before actual vasopressor initiation, during which time a median of 2,250 (IQR 1,200-3,300) mL of IVF was administered. This single-center analysis shows the feasibility of a quantitative, objective tool for describing usual care. Dynamic practice models may help assess when management was atypical; such tools may also be useful for designing and interpreting clinical trials.

Is novel coronavirus 2019 reinfection possible? Interpreting dynamic SARS-CoV-2 test results.

Since December 2019, COVID-19, the clinical syndrome associated with SARS-CoV-2 infection, has infected more than 6.2 million people and brought the function of the global community to a halt. As the number of patients recovered from COVID-19 rises and the world transitions toward reopening, the question of acquired immunity versus the possibility of reinfection are critical to anticipating future viral spread. Here, we present a case of a patient previously recovered from COVID-19 who re-presents with new respiratory, radiographical, laboratory, and real-time reverse transcriptase-polymerase chain reaction (RT-PCR) findings concerning for possible re-infection. We review this case in the context of the evolving discussion and theories surrounding dynamic RT-PCR results, prolonged viral shedding, and the possibility of developed immunity. Understanding how to interpret dynamic and late-positive SARS-CoV-2 RT-PCR results after primary infection will be critical for understanding disease prevalence and spread among communities worldwide.

Validation of a 'Usual Care' Model for Vasopressor Initiation in a Cohort of Emergency Department Patients with Sepsis.

Usual care regarding vasopressor (VP) initiation is ill-defined. We aimed to further validate a quantitative model for usual care in the Emergency Department (ED) regarding the timing of VP initiation in sepsis. We retrospectively studied a cohort of adult critically-ill ED patients who also received antibiotics in the ED. We applied a multivariable model previously developed from another patient cohort which distinguishes between time points at which patients were or were not subsequently started on a continuous VP infusion. The model has six independently significant predictors (respiratory rate, Glasgow Coma Scale score, systolic blood pressure, SpO2, administered intravenous fluids, and elapsed time). The outcome was initiation of VP infusion, either within the ED or within 6 hours after leaving the ED. We applied the model to all time points, beginning when all model input parameters were first available for a given patient, and ending when either VP were first started, or the patient left the ED. Out of 55,963 adult ED patients during the two-year study interval, we identified 1,629 who met our inclusion criteria. The area under the receiver operating characteristic curve was 0.81 for all patients, and 0.72 for the subset with at least one hypotensive blood pressure measurement. At a model threshold with sensitivity and specificity 0.74 and 0.74, respectively, the median advance detection time was 170.5 minutes (IQR 53 - 363).

Development and Validation of a Predictive Model for Hemodynamic Responses to Resuscitation during Uncontrolled Hemorrhage.

We investigated whether a statistical model could predict mean arterial pressure (MAP) during uncontrolled hemorrhage; such a model could be used for automated decision support, to help clinicians decide when to provide intravascular volume to achieve MAP goals. This was a secondary analysis of adult swine subjects during uncontrolled splenic bleeding. By protocol, after developing severe hypotension (MAP < 60 mmHg), subjects were resuscitated with either saline (NS) or fresh frozen plasma (FFP), determined randomly. Vital signs were documented at quasi-regular time-step intervals, until either subject death or 300 min. Subjects were randomly separated 50%/50% into training/validation sets, and regression models were developed to predict MAP for each subsequent (i.e., future) time-step. Median time-steps for serially recorded vital signs were +15 min. 5 subjects survived the protocol; 17 died after a median time of 87 min (IQR 78 - 134). The final model consisted of: current MAP; heart rate (HR); prior NS; imminent NS; and imminent FFP. The 95% limits-of-agreement between true subsequent MAP vs. predicted subsequent MAP were +10/-11 mmHg for the 79 time-steps in the training set; and +14/-13 for the 64 time-steps in the validation set. A total of 10 sudden death events (i.e., rapid, fatal MAP decrease within one single time-step) were excluded from analysis. In conclusion, for uncontrolled hemorrhage in a swine model, it was possible to estimate the next documented MAP value on the basis of the subject's current documented MAP; HR; prior NS; and the volume of resuscitation about to be administered. However, the model was unable to predict "sudden death" events. The applicability to populations with wider heterogeneity of hemorrhage patterns and with comorbidities requires further investigation.

Antibiotic Delays and Feasibility of a 1-Hour-From-Triage Antibiotic Requirement: Analysis of an Emergency Department Sepsis Quality Improvement Database.

STUDY OBJECTIVE: We identify factors associated with delayed emergency department (ED) antibiotics and determine feasibility of a 1-hour-from-triage antibiotic requirement in sepsis. METHODS: We studied all ED adult septic patients in accordance with Centers for Medicare & Medicaid Services Severe Sepsis and Septic Shock National Quality Measures in 2 consecutive 12-month intervals. During the second interval, a quality improvement intervention was conducted: a sepsis screening protocol plus case-specific feedback to clinicians. Data were abstracted retrospectively through electronic query and chart review. Primary outcomes were antibiotic delay greater than 3 hours from documented onset of hypoperfusion (per Centers for Medicare & Medicaid Services Severe Sepsis and Septic Shock National Quality Measures) and antibiotic delay greater than 1 hour from triage (per 2018 Surviving Sepsis Campaign recommendations). RESULTS: We identified 297 and 357 septic patients before and during the quality improvement intervention, respectively. Before and during quality improvement intervention, antibiotic delay in accordance with Centers for Medicare & Medicaid Services measures occurred in 30% and 21% of cases (-9% [95% confidence interval -16% to -2%]); and in accordance with 2018 Surviving Sepsis Campaign recommendations, 85% and 71% (-14% [95% confidence interval -20% to -8%]). Four factors were independently associated with both definitions of antibiotic delay: vague (ie, nonexplicitly infectious) presenting symptoms, triage location to nonacute areas, care before the quality improvement intervention, and lower Sequential [Sepsis-related] Organ Failure Assessment scores. Most patients did not receive antibiotics within 1 hour of triage, with the exception of a small subset post-quality improvement intervention who presented with explicit infectious symptoms and triage hypotension. CONCLUSION: The quality improvement intervention significantly reduced antibiotic delays, yet most septic patients did not receive antibiotics within 1 hour of triage. Compliance with the 2018 Surviving Sepsis Campaign would require a wholesale alteration in the management of ED patients with either vague symptoms or absence of triage hypotension.

Clustering Blood Pressure Trajectories in Septic Shock in the Emergency Department.

Hemodynamic management of sepsis in the emergency department relies on fluid resuscitation and vasoactive therapy to maintain adequate blood pressure and end-organ perfusion. While typical practice targets certain thresholds of blood pressure (such as 65 mmHg mean arterial or 90 mmHg systolic blood pressure [SBP]), little consideration is given to temporal dynamics of blood pressure. In this work, we use unsupervised learning methods to reveal characteristic SBP trajectories in the two hours either surrounding the start of hypotensive episodes (SBP <; 90 mmHg) or immediately preceding the initiation of vasopressor therapy. Our results show that hypotensive episodes tended to either resolve very quickly (within 40 minutes) or extend for prolonged periods (at least 1 hour). Those with prolonged hypotension constituted 74% of all patients with at least one measurement of SBP <; 90 mmHg. Of them, patients who entered hypotension by a large, acute drop from a normal SBP over the preceding hour had a greater incidence of subsequent vasopressor administration than those with a more gradual decline into hypotension. Overall, our results suggest that a significant subset of patients, especially those with stable but low SBP, should have received vasopressors when they did not, or should have received them sooner. Dynamic trajectories appear to be important factors in clinical practice of hemodynamic management of sepsis.

Forecasting Hypotension during Vasopressor Infusion via Time Series Analysis.

For optimal management of hypotension during continuous vasopressor infusion, this study investigated two forecasting models, logistic regression (LR) and auto-regressive (AR) models, to predict sustained hypotension episodes (SHEs) in the ICU, before the SHE occurred. Two investigational models were compared to a simple threshold detector, which alerts whenever the BP is less than the specific hypotension threshold. Datasets were collected from 207 patients treated for a variety of clinical indications in two different hospitals (Hospital 1 & 2). For the 60 mmHg hypotension threshold, LR model predicted SHEs an average of 7.0 min before (Hospital 1) and 2.5 min before (Hospital 2), and the AR model predicted SHEs 10.5 min and 2.0 min before (Hospital 1 and 2 respectively). Both were significantly better than the threshold method and without higher false alarm rates. The AR model offered the flexibility to predict for different hypotension thresholds.

Repeat CT Performed Within One Month of CT Conducted in the Emergency Department for Abdominal Pain: A Secondary Analysis of Data From a Prospective Multicenter Study.

OBJECTIVE: The purpose of this study is to determine both the frequency of repeat CT performed within 1 month after a patient visits the emergency department (ED) and undergoes CT evaluation for abdominal pain and the frequency of worsened or new CT-based diagnoses. SUBJECTS AND METHODS: Secondary analysis was performed on data collected during a prospective multicenter study. The parent study included patients who underwent CT in the ED for abdominal pain between 2012 and 2014, and these patients constituted the study group of the present analysis. The proportion of patients who underwent (in any setting) repeat abdominal CT within 1 month of the index CT examination was calculated. For each of these patients, results of the index and repeat CT scans were compared by an independent panel and categorized as follows: no change (group 1); same process, improved (group 2); same process, worse (group 3); or different process (group 4). The proportion of patients in groups 1 and 2 versus groups 3 and 4 was calculated, and patient and ED physician characteristics were compared. RESULTS: The parent study included 544 patients (246 of whom were men [45%]; mean patient age, 49.4 years). Of those 544 patients, 53 (10%; 95% CI, 7.5-13%) underwent repeat abdominal CT. Patients' CT comparisons were categorized as follows: group 1 for 43% of patients (23/53), group 2 for 26% (14/53), group 3 for 15% (8/53), and group 4 for 15% (8/53). New or worse findings were present in 30% of patients (16/53) (95% CI, 19-44%). When patients with findings in groups 1 and 2 were compared to patients with findings in groups 3 and 4, no significant difference was noted in patient age (p = 0.25) or sex (p = 0.76), the number of days between scans (p = 0.98), and the diagnostic confidence of the ED physician after the index CT scan was obtained (p = 0.33). CONCLUSION: Short-term, repeat abdominal CT was performed for 10% of patients who underwent CT in the ED for abdominal pain, and it yielded new or worse findings for 30% of those patients.

Challenges and Opportunities for Emergency Department Sepsis Screening at Triage.

Feasibility of ED triage sepsis screening, before diagnostic testing has been performed, has not been established. In a retrospective, outcome-blinded chart review of a one-year cohort of ED adult septic shock patients ("derivation cohort") and three additional, non-consecutive months of all adult ED visits ("validation cohort"), we evaluated the qSOFA score, the Shock Precautions on Triage (SPoT) vital-signs criterion, and a triage concern-for-infection (tCFI) criterion based on risk factors and symptoms, to screen for sepsis. There were 19,670 ED patients in the validation cohort; 50 developed ED septic shock, of whom 60% presented without triage hypotension, and 56% presented with non-specific symptoms. The tCFI criterion improved specificity without substantial reduction of sensitivity. At triage, sepsis screens (positive qSOFA vital-signs and tCFI, or positive SPoT vital-signs and tCFI) were 28% (95% CI: 16-43%) and 56% (95% CI: 41-70%) sensitive, respectively, p < 0.01. By the conclusion of the ED stay, sensitivities were 80% (95% CI: 66-90%) and 90% (95% CI: 78-97%), p > 0.05, and specificities were 97% (95% CI: 96-97%) and 95% (95% CI: 95-96%), p < 0.001. ED patients who developed septic shock requiring vasopressors often presented normotensive with non-specific complaints, necessitating a low threshold for clinical concern-for-infection at triage.

Presenting Symptoms Independently Predict Mortality in Septic Shock: Importance of a Previously Unmeasured Confounder.

OBJECTIVES: Presenting symptoms in patients with sepsis may influence rapidity of diagnosis, time-to-antibiotics, and outcome. We tested the hypothesis that vague presenting symptoms are associated with delayed antibiotics and increased mortality. We further characterized individual presenting symptoms and their association with mortality. DESIGN: Retrospective cohort study. SETTING: Emergency department of large, urban, academic U.S. hospital. PATIENTS: All adult patients with septic shock treated in the emergency department between April 2014 and March 2016. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Of 654 septic shock cases, 245 (37%) presented with vague symptoms. Time-to-antibiotics from first hypotension or elevated lactate was significantly longer for those with vague symptoms versus those with explicit symptoms of infection (1.6 vs 0.8 hr; p < 0.01), and in-hospital mortality was also substantially higher (34% vs 16%; p < 0.01). Patients with vague symptoms were older and sicker as evidenced by triage hypotension, Sequential Organ Failure Assessment score, initial serum lactate, and need for intubation. In multivariate analysis, vague symptoms were independently associated with mortality (adjusted odds ratio, 2.12; 95% CI, 1.32-3.40; p < 0.01), whereas time-to-antibiotics was not associated with mortality (adjusted odds ratio, 1.01; 95% CI, 0.94-1.08; p = 0.78). Of individual symptoms, only the absence of fever, chills, or rigors (odds ratio, 2.70; 95% CI, 1.63-4.47; p < 0.01) and presence of shortness of breath (odds ratio, 1.97; 95% CI, 1.23-3.15; p < 0.01) were independently associated with mortality. CONCLUSIONS: More than one third of patients with septic shock presented to the emergency department with vague symptoms that were not specific to infection. These patients had delayed antibiotic administration and higher risk of mortality even after controlling for demographics, illness acuity, and time-to-antibiotics in multivariate analysis. These findings suggest that the nature of presenting symptoms is an important component of sepsis clinical phenotyping and may be an important confounder in sepsis epidemiologic studies.