Phenotypes of Patients with Intracerebral Hemorrhage, Complications, and Outcomes.

BACKGROUND: The objective of this study was to define clinically meaningful phenotypes of intracerebral hemorrhage (ICH) using machine learning. METHODS: We used patient data from two US medical centers and the Antihypertensive Treatment of Acute Cerebral Hemorrhage-II clinical trial. We used k-prototypes to partition patient admission data. We then used silhouette method calculations and elbow method heuristics to optimize the clusters. Associations between phenotypes, complications (e.g., seizures), and functional outcomes were assessed using the Kruskal-Wallis H-test or χ2 test. RESULTS: There were 916 patients; the mean age was 63.8 ± 14.1 years, and 426 patients were female (46.5%). Three distinct clinical phenotypes emerged: patients with small hematomas, elevated blood pressure, and Glasgow Coma Scale scores > 12 (n = 141, 26.6%); patients with hematoma expansion and elevated international normalized ratio (n = 204, 38.4%); and patients with median hematoma volumes of 24 (interquartile range 8.2-59.5) mL, who were more frequently Black or African American, and who were likely to have intraventricular hemorrhage (n = 186, 35.0%). There were associations between clinical phenotype and seizure (P = 0.024), length of stay (P = 0.001), discharge disposition (P < 0.001), and death or disability (modified Rankin Scale scores 4-6) at 3-months' follow-up (P < 0.001). We reproduced these three clinical phenotypes of ICH in an independent cohort (n = 385) for external validation. CONCLUSIONS: Machine learning identified three phenotypes of ICH that are clinically significant, associated with patient complications, and associated with functional outcomes. Cerebellar hematomas are an additional phenotype underrepresented in our data sources.

Precise Image-level Localization of Intracranial Hemorrhage on Head CT Scans with Deep Learning Models Trained on Study-level Labels.

"Just Accepted" papers have undergone full peer review and have been accepted for publication in Radiology: Artificial Intelligence. This article will undergo copyediting, layout, and proof review before it is published in its final version. Please note that during production of the final copyedited article, errors may be discovered which could affect the content. Purpose To develop a highly generalizable weakly supervised model to automatically detect and localize image- level intracranial hemorrhage (ICH) using study-level labels. Materials and Methods In this retrospective study, the proposed model was pretrained on the image-level RSNA dataset and fine-tuned on a local dataset using attention-based bidirectional long-short-term memory networks. This local training dataset included 10,699 noncontrast head CT scans from 7469 patients with ICH study-level labels extracted from radiology reports. Model performance was compared with that of two senior neuroradiologists on 100 random test scans using the McNemar test, and its generalizability was evaluated on an external independent dataset. Results The model achieved a positive predictive value (PPV) of 85.7% (95% CI: [84.0%, 87.4%]) and an AUC of 0.96 (95% CI: [0.96, 0.97]) on the held-out local test set (n = 7243, 3721 female) and 89.3% (95% CI: [87.8%, 90.7%]) and 0.96 (95% CI: [0.96, 0.97]), respectively, on the external test set (n = 491, 178 female). For 100 randomly selected samples, the model achieved performance on par with two neuroradiologists, but with a significantly faster (P < .05) diagnostic time of 5.04 seconds per scan (versus 86 seconds and 22.2 seconds for the two neuroradiologists, respectively). The model's attention weights and heatmaps visually aligned with neuroradiologists' interpretations. Conclusion The proposed model demonstrated high generalizability and high PPVs, offering a valuable tool for expedited ICH detection and prioritization while reducing false-positive interruptions in radiologists' workflows. ©RSNA, 2024.

Northwestern University resource and education development initiatives to advance collaborative artificial intelligence across the learning health system.

Introduction: The rapid development of artificial intelligence (AI) in healthcare has exposed the unmet need for growing a multidisciplinary workforce that can collaborate effectively in the learning health systems. Maximizing the synergy among multiple teams is critical for Collaborative AI in Healthcare. Methods: We have developed a series of data, tools, and educational resources for cultivating the next generation of multidisciplinary workforce for Collaborative AI in Healthcare. We built bulk-natural language processing pipelines to extract structured information from clinical notes and stored them in common data models. We developed multimodal AI/machine learning (ML) tools and tutorials to enrich the toolbox of the multidisciplinary workforce to analyze multimodal healthcare data. We have created a fertile ground to cross-pollinate clinicians and AI scientists and train the next generation of AI health workforce to collaborate effectively. Results: Our work has democratized access to unstructured health information, AI/ML tools and resources for healthcare, and collaborative education resources. From 2017 to 2022, this has enabled studies in multiple clinical specialties resulting in 68 peer-reviewed publications. In 2022, our cross-discipline efforts converged and institutionalized into the Center for Collaborative AI in Healthcare. Conclusions: Our Collaborative AI in Healthcare initiatives has created valuable educational and practical resources. They have enabled more clinicians, scientists, and hospital administrators to successfully apply AI methods in their daily research and practice, develop closer collaborations, and advanced the institution-level learning health system.

P2Y12 inhibitor use predicts hematoma expansion in patients with intracerebral hemorrhage.

OBJECTIVE: Hematoma expansion (HE) predicts disability and death after acute intracerebral hemorrhage (ICH). Aspirin and anticoagulants have been associated with HE. We tested the hypothesis that P2Y12 inhibitors predict subsequent HE in patients. We explored laboratory measures of P2Y12 inhibition and dual antiplatelet therapy with aspirin (DAPT). METHODS: We prospectively identified patients with ICH. Platelet activity was measured with the VerifyNow-P2Y12 assay. Hematoma volumes for initial and follow-up CTs were calculated using a validated semi-automated technique. HE was defined as the difference between hematoma volumes on the initial and follow-up CT scans. Nonparametric statistics were performed with Kruskal-Wallis H, and correction for multiple comparisons performed with Dunn's test. RESULTS: In 194 patients, 15 (7.7%) were known to take a P2Y12 inhibitor (clopidogrel in all but one). Patients taking a P2Y12 inhibitor had more HE compared to patients not taking a P2Y12 inhibitor (3.5 [1.2-11.9] vs. 0.1 [-0.8-1.4] mL, p = 0.004). Patients taking DAPT experienced the most HE (7.2 [2.6-13.8] vs. 0.0 [-1.0-1.1] mL, p = 0.04). The use of P2Y12 inhibitors was associated with less P2Y12 activity (178 [149-203] vs. 288 [246-319] P2Y12 reaction units, p = 0.005). INTERPRETATION: Patients taking a P2Y12 inhibitor had more HE and less P2Y12 activity. The effect was most pronounced in patients on DAPT, suggesting a synergistic effect of P2Y12 inhibitors and aspirin with respect to HE. Acute reversal of P2Y12 inhibitors in acute ICH requires further study.

Fixed- Versus Variable-Dose Prothrombin Complex Concentrate for the Emergent Reversal of Vitamin K Antagonists: A Systematic Review and Meta-Analysis.

OBJECTIVES: Four-factor prothrombin complex concentrate (4-PCC) is recommended for rapid reversal of vitamin K antagonists (VKAs) such as warfarin, yet optimal dosing remains uncertain. DATA SOURCES: A systematic review was conducted of PubMed, Embase, and Ovid MEDLINE (Wolters Kluwer) databases from January 2000 to August 2023 for clinical studies comparing fixed- vs. variable-dose 4-PCC for emergent VKA reversal with at least one reported clinical outcome. STUDY SELECTION: Abstracts and full texts were assessed independently and in duplicate by two reviewers. DATA EXTRACTION: Data were extracted independently and in duplicate by two reviewers using predefined extraction forms. DATA SYNTHESIS: The analysis comprised three randomized trials and 16 cohort studies comprising a total of 323 participants in randomized trials (161 in fixed dosage and 162 in variable dosage) and 1912 patients in cohort studies (858 in fixed-dose and 1054 in variable dose). Extracranial bleeding was the predominant indication, while intracranial hemorrhage varied. Overall, a fixed-dose regimen may be associated with a lower dose of 4-PCC and results in a reduction in 4-PCC administration time compared with a variable-dose regimen. A fixed-dose regimen also likely results in increased clinical hemostasis. While there is no clear difference between the two regimens in terms of achieving a goal international normalized ratio (INR) less than 2, a fixed-dose regimen is less likely to achieve a goal INR less than 1.5. High certainty evidence indicates that the fixed-dose regimen reduces both mortality and the occurrence of thromboembolic events. Additional subgroup analyses provides exploratory data to guide future studies. CONCLUSIONS: A fixed-dose regimen for 4-PCC administration provides benefits over a variable-dose regimen in terms of dose reduction, faster administration time, improved clinical hemostasis, and reduced mortality and thromboembolic events. Further studies are warranted to better refine the optimal fixed-dose regimen.

External Validation of a Tool to Identify Low-Risk Patients With Isolated Subdural Hematoma and Preserved Consciousness.

STUDY OBJECTIVE: The SafeSDH Tool was derived to identify patients with isolated (no other type of intracranial hemorrhage) subdural hematoma who are at very low risk of neurologic deterioration, neurosurgical intervention, or death. Patients are low risk by the tool if they have none of the following: use of anticoagulant or nonaspirin antiplatelet agent, Glasgow Coma Score (GCS) <14, more than 1 discrete hematoma, hematoma thickness >5 mm, or midline shift. We attempted to externally validate the SafeSDH Tool. METHODS: We performed a retrospective chart review of patients aged ≥16 with a GCS ≥13 and isolated subdural hematoma who presented to 1 of 6 academic and community hospitals from 2005 to 2018. The primary outcome, a composite of neurologic deterioration (seizure, altered mental status, or symptoms requiring repeat imaging), neurosurgical intervention, discharge on hospice, and death, was abstracted from discharge summaries. Hematoma thickness, number of hematomas, and midline shift were abstracted from head imaging reports. Anticoagulant use, antiplatelet use, and GCS were gathered from the admission record. RESULTS: The validation data set included 753 patients with isolated subdural hematoma. Mortality during the index admission was 2.1%; 26% of patients underwent neurosurgical intervention. For the composite outcome, sensitivity was 99% (95% confidence interval [CI] 97 to 100), and specificity was 31% (95% CI 27 to 35). The tool identified 162 (21.5%) patients as low risk. Negative likelihood ratio was 0.03 (95% CI 0.01 to 0.11). CONCLUSION: The SafeSDH Tool identified patients with isolated subdural hematoma who are at low risk for poor outcomes with high sensitivity. With prospective validation, these low-risk patients could be safe for management in less intensive settings.

Blood pressure medication and acute kidney injury after intracerebral haemorrhage: an analysis of the ATACH-II trial.

Background: Acute blood pressure (BP) reduction is standard of care after acute intracerebral haemorrhage (ICH). More acute BP reduction is associated with acute kidney injury (AKI). It is not known if the choice of antihypertensive medications affects the risk of AKI. Methods: We analysed data from the ATACH-II clinical trial. AKI was defined by the Kidney Disease: Improving Global Outcomes criteria. We analysed antihypertensive medication from two sources. The first was a case report form that specified the use of labetalol, diltiazem, urapidil or other. We tested the hypothesis that the secondary medication was associated with AKI with χ2 test. Second, we tested the hypotheses the dosage of diltiazem was associated with AKI using Mann-Whitney U test. Results: AKI occurred in 109 of 1000 patients (10.9%). A higher proportion of patients with AKI received diltiazem after nicardipine (12 (29%) vs 21 (12%), p=0.03). The 95%ile (90%-99% ile) of administered diltiazem was 18 (0-130) mg in patients with AKI vs 0 (0-30) mg in patients without AKI (p=0.002). There was no apparent confounding by indication for diltiazem use. Conclusions: The use of diltiazem, and more diltiazem, was associated with AKI in patients with acute ICH.

Big Data in Stroke: How to Use Big Data to Make the Next Management Decision.

The last decade has seen significant advances in the accumulation of medical data, the computational techniques to analyze that data, and corresponding improvements in management. Interventions such as thrombolytics and mechanical thrombectomy improve patient outcomes after stroke in selected patients; however, significant gaps remain in our ability to select patients, predict complications, and understand outcomes. Big data and the computational methods needed to analyze it can address these gaps. For example, automated analysis of neuroimaging to estimate the volume of brain tissue that is ischemic and salvageable can help triage patients for acute interventions. Data-intensive computational techniques can perform complex risk calculations that are too cumbersome to be completed by humans, resulting in more accurate and timely prediction of which patients require increased vigilance for adverse events such as treatment complications. To handle the accumulation of complex medical data, a variety of advanced computational techniques referred to as machine learning and artificial intelligence now routinely complement traditional statistical inference. In this narrative review, we explore data-intensive techniques in stroke research, how it has informed the management of stroke patients, and how current work could shape clinical practice in the future.

Patterns of diverse and changing sentiments towards COVID-19 vaccines: a sentiment analysis study integrating 11 million tweets and surveillance data across over 180 countries.

OBJECTIVES: Vaccines are crucial components of pandemic responses. Over 12 billion coronavirus disease 2019 (COVID-19) vaccines were administered at the time of writing. However, public perceptions of vaccines have been complex. We integrated social media and surveillance data to unravel the evolving perceptions of COVID-19 vaccines. MATERIALS AND METHODS: Applying human-in-the-loop deep learning models, we analyzed sentiments towards COVID-19 vaccines in 11 211 672 tweets of 2 203 681 users from 2020 to 2022. The diverse sentiment patterns were juxtaposed against user demographics, public health surveillance data of over 180 countries, and worldwide event timelines. A subanalysis was performed targeting the subpopulation of pregnant people. Additional feature analyses based on user-generated content suggested possible sources of vaccine hesitancy. RESULTS: Our trained deep learning model demonstrated performances comparable to educated humans, yielding an accuracy of 0.92 in sentiment analysis against our manually curated dataset. Albeit fluctuations, sentiments were found more positive over time, followed by a subsequence upswing in population-level vaccine uptake. Distinguishable patterns were revealed among subgroups stratified by demographic variables. Encouraging news or events were detected surrounding positive sentiments crests. Sentiments in pregnancy-related tweets demonstrated a lagged pattern compared with the general population, with delayed vaccine uptake trends. Feature analysis detected hesitancies stemmed from clinical trial logics, risks and complications, and urgency of scientific evidence. DISCUSSION: Integrating social media and public health surveillance data, we associated the sentiments at individual level with observed populational-level vaccination patterns. By unraveling the distinctive patterns across subpopulations, the findings provided evidence-based strategies for improving vaccine promotion during pandemics.

Fall Risk Prediction Models During the Initial COVID-19 Surge: Could Predictive Analytics Be Used in a Resource-Constrained Environment?

During the first COVID surge, multiple changes in nurse staffing and workflows were made to support care delivery in a resource-constrained environment. We hypothesized that there was a higher rate of inpatient falls during the COVID surge. Furthermore, we predicted that an automated predictive analytic algorithm would perform as well as the Johns Hopkins Fall Risk Assessment. A retrospective review of falls for 3 months before and the first 3 months of the first COVID surge was conducted. We determined the total number of falls and the overall fall rate and examined the distribution of scores and accuracy of fall predictive models for both groups. There was a statistically significant increase in fall rate during the first 3 months of the COVID surge compared with the 3 prior months (2.48/1000 patient-days vs 1.89/1000 patient-days respectively; P = .041). The Johns Hopkins instrument had a greater sensitivity of 78.9% compared with 57.0% for the predictive analytic model. Specificity and accuracy of the predictive analytic model were higher than the Johns Hopkins instrument (71.3% vs 54.1% and 71.2% vs 54.3%, respectively). These findings suggest that the automated predictive analytic model could be used in a resource-constrained environment to accurately classify patients' risk of fall.

Effect of Magnesium on Deterioration and Symptomatic Hemorrhagic Transformation in Cerebral Ischemia: An Ancillary Analysis of the FAST-MAG Trial.

BACKGROUND: Magnesium (Mg) is a neuroprotectant in preclinical models. Lower serum Mg levels have been associated with symptomatic hemorrhagic transformation (HT) in patients with ischemic stroke. Early treatment of acute ischemic stroke with Mg may reduce rates of symptomatic HT. METHODS: In this post hoc study of the Field Administration of Stroke Therapy Magnesium (FAST-MAG) trial, 1,245 participants with a diagnosis of cerebral ischemia received 20 g of Mg or placebo initiated in the prehospital setting. Posttreatment serum Mg level was measured for 809 participants. Cases of clinical deterioration, defined as worsening by ≥4 points on the National Institute of Health Stroke Scale (NIHSS), were imaged and evaluated for etiology. Symptomatic HT was defined as deterioration with imaging showing new hemorrhage. RESULTS: Clinical deterioration occurred in 187 and symptomatic HT in 46 of 1,245 cases of cerebral ischemia. Rates of deterioration and symptomatic HT were not significantly lower in those who received Mg (15.7% vs. 14.4%, p = 0.591; 2.8% vs. 4.6%, p = 0.281). In cases where serum Mg level was obtained posttreatment, lower serum Mg level (<1.7 mg/dL) was associated with significantly higher rates of deterioration and symptomatic HT (27.5% vs. 15.5%, p = 0.0261; 11.6% vs. 3.65%, p = 0.00819). CONCLUSIONS: Treatment with Mg did not significantly reduce rates of clinical deterioration or symptomatic HT. Future analysis should address whether treatment with Mg could have influenced the subgroup with low serum Mg at baseline.

Recruitment in Acute Stroke Trials: Challenges and Potential Solutions.

Randomized clinical trials of acute stroke have led to major advances in acute stroke therapy over the past decade. Despite these successes, recruitment in acute trials is often difficult. We outline challenges in recruitment for acute stroke trials and present potential solutions, which can increase the speed and decrease the cost of identifying new treatments for acute stroke. One of the largest opportunities to increase the speed of enrollment and make trials more generalizable is expansion of inclusion criteria whose impact on expected recruitment can be assessed by epidemiologic and registry databases. Another barrier to recruitment besides the number of eligible patients is availability of study investigators limited to business hours, which may be helped by financial support for after-hours call. The wider use of telemedicine has accelerated quicker stroke treatment at many hospitals and has the potential to accelerate research enrollment but requires training of clinical investigators who are often inexperienced with this approach. Other potential solutions to enhance recruitment include rapid prehospital notification of clinical investigators of potential patients, use of mobile stroke units, advances in the process of emergency informed consent, storage of study medication in the emergency department, simplification of study treatments and data collection, education of physicians to improve equipoise and enthusiasm for randomization of patients within a trial, and clear recruitment plans, and even potentially coenrollment, when there are competing trials at sites. Without successful recruitment, scientific advances and clinical benefit for acute stroke patients will lag.

Comparison between machine learning methods for mortality prediction for sepsis patients with different social determinants.

BACKGROUND: Sepsis is one of the most life-threatening circumstances for critically ill patients in the United States, while diagnosis of sepsis is challenging as a standardized criteria for sepsis identification is still under development. Disparities in social determinants of sepsis patients can interfere with the risk prediction performances using machine learning. METHODS: We analyzed a cohort of critical care patients from the Medical Information Mart for Intensive Care (MIMIC)-III database. Disparities in social determinants, including race, sex, marital status, insurance types and languages, among patients identified by six available sepsis criteria were revealed by forest plots with 95% confidence intervals. Sepsis patients were then identified by the Sepsis-3 criteria. Sixteen machine learning classifiers were trained to predict in-hospital mortality for sepsis patients on a training set constructed by random selection. The performance was measured by area under the receiver operating characteristic curve (AUC). The performance of the trained model was tested on the entire randomly conducted test set and each sub-population built based on each of the following social determinants: race, sex, marital status, insurance type, and language. The fluctuations in performances were further examined by permutation tests. RESULTS: We analyzed a total of 11,791 critical care patients from the MIMIC-III database. Within the population identified by each sepsis identification method, significant differences were observed among sub-populations regarding race, marital status, insurance type, and language. On the 5783 sepsis patients identified by the Sepsis-3 criteria statistically significant performance decreases for mortality prediction were observed when applying the trained machine learning model on Asian and Hispanic patients, as well as the Spanish-speaking patients. With pairwise comparison, we detected performance discrepancies in mortality prediction between Asian and White patients, Asians and patients of other races, as well as English-speaking and Spanish-speaking patients. CONCLUSIONS: Disparities in proportions of patients identified by various sepsis criteria were detected among the different social determinant groups. The performances of mortality prediction for sepsis patients can be compromised when applying a universally trained model for each subpopulation. To achieve accurate diagnosis, a versatile diagnostic system for sepsis is needed to overcome the social determinant disparities of patients.

Magnesium Sulfate and Hematoma Expansion: An Ancillary Analysis of the FAST-MAG Randomized Trial.

BACKGROUND: Intracerebral hemorrhage (ICH) is the deadliest form of stroke. In observational studies, lower serum magnesium has been linked to more hematoma expansion (HE) and intracranial hemorrhage, implying that supplemental magnesium sulfate is a potential acute treatment for patients with ICH and could reduce HE. FAST-MAG (Field Administration of Stroke Therapy - Magnesium) was a clinical trial of magnesium sulfate started prehospital in patients with acute stroke within 2 hours of last known well enrolled. CT was not required prior to enrollment, and several hundred patients with acute ICH were enrolled. In this ancillary analysis, we assessed the effect of magnesium sulfate treatment upon HE in patients with acute ICH. METHODS: We retrospectively analyzed data that were prospectively collected in the FAST-MAG study. Patients received intravenous magnesium sulfate or matched placebo within 2 hours of onset. We compared HE among patients allocated to intravenous magnesium sulfate or placebo with a Mann-Whitney U. We used the same method to compare neurological deficit severity (National Institutes of Health Stroke Scale) and global disability (modified Rankin Scale) at 3 months. RESULTS: Among 268 patients with ICH meeting study entry criteria, mean 65.4±13/4 years, 33% were female, and 211 (79%) had a history of hypertension. Initial deficit severities were median (interquartile range) of 4 (3-5) on the Los Angeles Motor Scale in the field and National Institutes of Health Stroke Scale score of 16 (9.5-25.5) early after hospital arrival. Follow-up brain imaging was performed a median of 17.1 (11.3-22.7) hours after first scan. The magnesium and placebo groups did not statistically differ in hematoma volume on arrival, 10.1 (5.6-28.7) versus 12.4 (5.6-28.7) mL (P=0.6), or HE, 2.0 (0.1-7.4) versus 1.5 (-0.2 to 8) mL (P=0.5). There was no difference in functional outcomes (modified Rankin Scale score of 3-6), 59% versus 50% (P=0.5). CONCLUSIONS: Magnesium sulfate did not reduce HE or improve functional outcomes at 90 days. A benefit for patients with initial hypomagnesemia was not addressed. REGISTRATION: URL: https://www. CLINICALTRIALS: gov; Unique identifier: NCT00059332.

Predicting Early Seizures After Intracerebral Hemorrhage with Machine Learning.

BACKGROUND: Seizures are a harmful complication of acute intracerebral hemorrhage (ICH). "Early" seizures in the first week after ICH are a risk factor for deterioration, later seizures, and herniation. Ideally, seizure medications after ICH would only be administered to patients with a high likelihood to have seizures. We developed and validated machine learning (ML) models to predict early seizures after ICH. METHODS: We used two large datasets to train and then validate our models in an entirely independent test set. The first model ("CAV") predicted early seizures from a subset of variables of the CAVE score (a prediction rule for later seizures)-cortical hematoma location, age less than 65 years, and hematoma volume greater than 10 mL-whereas early seizure was the dependent variable. We attempted to improve on the "CAV" model by adding anticoagulant use, antiplatelet use, Glasgow Coma Scale, international normalized ratio, and systolic blood pressure ("CAV + "). For each model we used logistic regression, lasso regression, support vector machines, boosted trees (Xgboost), and random forest models. Final model performance was reported as the area under the receiver operating characteristic curve (AUC) using receiver operating characteristic models for the test data. The setting of the study was two large academic institutions: institution 1, 634 patients; institution 2, 230 patients. There were no interventions. RESULTS: Early seizures were predicted across the ML models by the CAV score in test data, (AUC 0.72, 95% confidence interval 0.62-0.82). The ML model that predicted early seizure better in the test data was Xgboost (AUC 0.79, 95% confidence interval 0.71-0.87, p = 0.04) compared with the CAV model AUC. CONCLUSIONS: Early seizures after ICH are predictable. Models using cortical hematoma location, age less than 65 years, and hematoma volume greater than 10 mL had a good accuracy rate, and performance improved with more independent variables. Additional methods to predict seizures could improve patient selection for monitoring and prophylactic seizure medications.

Antiplatelet Medications and Biomarkers of Hemostasis May Explain the Association of Hematoma Appearance and Subsequent Hematoma Expansion After Intracerebral Hemorrhage.

BACKGROUND: To test the hypothesis that appearances of intracranial hematomas on diagnostic computed tomography (CT) are not idiosyncratic and reflect a biologically plausible mechanism, we evaluated the association between hematoma appearance on CT, biomarkers of platelet activity, and antiplatelet or anticoagulant medication use prior to admission. METHODS: We studied 330 consecutively identified patients from 2006 to 2019. Biomarkers of platelet activity (platelet aspirin assay) and medication history (aspirin, clopidogrel) were prospectively recorded on admission. A blinded interpreter recorded the presence of hematoma appearances from the diagnostic scan. Associations were tested with parametric or nonparametric statistics, as appropriate. RESULTS: The black hole sign (101, 30%) was most prevalent, followed by the island sign (57, 17%) and blend sign (32, 10%). There was reduced platelet activity in patients with a black hole sign (511 [430-610] vs. 562 [472-628] aspirin reaction units, P = 0.01) or island sign (505 [434-574] vs. 559 [462-629] aspirin reaction units, P = 0.004). Clopidogrel use prior to admission was associated with the black hole sign (odds ratio 2.25, 95% confidence interval 1.02-4.98, P = 0.04). CONCLUSIONS: In patients with acute intracerebral hemorrhage, hematoma appearances on CT are associated with biomarkers of platelet activity and clopidogrel use prior to admission. Appearances of intracranial hematomas on CT may reflect reduced hemostasis from antiplatelet medication use.

Comparison of Two Different Models to Predict Fall Risk in Hospitalized Patients.

BACKGROUND: Fall prevention is a patient safety and economic priority for health care organizations. An automated model within the electronic medical record (EMR) that accurately predicts risk for falling would be valuable for mitigation of inpatient falls. The aim of this study was to validate the reliability of an EMR-based computerized predictive model (ROF Model) for inpatient falls. The hypothesis was that the ROF Model would be similar to the Johns Hopkins Fall Risk Assessment Tool (JHFRAT) in predicting fall events in the inpatient setting at a large academic medical center. METHODS: This observational study compared the falls predicted by each model against actual falls over an eight-month period in a single institution. Descriptive statistics were used to compare the distribution of scores and accuracy of fall risk categorization for each model immediately preceding a fall. RESULTS: For 35,709 inpatient encounters, the total fall rate was 0.92%. Of the 329 patients who fell, 60.8% were high risk by ROF Model (fall rate 1.82%), and 75.4% were high risk by JHFRAT (fall rate 1.39%). The ROF Model had a better specificity than the JHFRAT (69.7% vs. 49.2%) but a similar C-statistic (0.717 vs. 0.702) and a lower sensitivity (60.8% vs. 79.3%). CONCLUSION: The performance of the ROF Model was similar to that of the JHFRAT in predicting inpatient falls. This comparison provides evidence to support a transition to a more automated process. Future studies will determine prospectively if implementation of the ROF Model will reduce falls in the inpatient setting.