Search

Variety of software solutions are being used for clinical use. This special contribution focuses on the personal experience of developing several software solutions concerning stroke. Stroke119 application was developed to inform the patient of the closest hospital available for thrombolytic therapy and provides a simple three-step self-test for detection of acute stroke. A multi-center web-based registry solution named SMART DB was developed to facilitate multi-center studies. Over 650,000 records were created by 25 centers in SMART DB. An artificial intelligence-based web solution for prediction of coronary artery disease in stroke patients was developed named S2CAD. A clinical decision support platform for thrombi acquired from endovascular thrombectomy named ARIA Cloud was developed. Software for stroke is actively being developed in Korea. Software solutions are expected to increase efficiency by providing clinical decision support in the near future.

Background and ObjectivesA machine learning technique for identifying hidden coronary artery disease (CAD) might be useful. We developed and validated machine learning models to predict patients with hidden CAD and to assess long-term outcomes in patients with acute ischemic stroke.MethodsMultidetector coronary CT was performed for patients without a known history of CAD. Primary outcomes were defined as having any degree of CAD and having obstructive CAD (≥50% stenosis). Demographic variables, risk factors, laboratory results, Trial of ORG 10172 in Acute Stroke Treatment classification, NIH Stroke Scale score, blood pressure, and carotid artery stenosis were used to develop and validate machine learning models to predict CAD. Area under the receiver operating characteristic curves (AUC) was calculated for performance analysis, and Kaplan-Meier and Cox survival analyses of long-term outcomes were performed. Major adverse cardiovascular events (MACEs) were defined as ischemic stroke, myocardial infarction, unstable angina, urgent coronary revascularization, and cardiovascular mortality.ResultsOverall, 1,710 patients were included for the training dataset and 348 patients for the validation dataset. An extreme gradient boosting model was developed to predict any degree of CAD, which showed an AUC of 0.763 (95% CI 0.711–0.814) on validation. A logistic regression model was used to predict obstructive CAD and had an AUC of 0.714 (95% CI 0.692–0.799). During the first 5 years of follow-up, MACEs occurred more frequently with predictions of any CAD (p = 0.022) or obstructive CAD (p < 0.001). Cox proportional analysis showed that the hazard ratio of MACE was 1.5 (95% CI 1.1–2.2; p = 0.016) with prediction of any CAD, whereas it was 1.9 (95% CI 1.3–2.6; p < 0.001) for obstructive CAD.DiscussionWe demonstrated that machine learning may help identify hidden CAD in patients with acute ischemic stroke. Long-term outcomes were also associated with prediction results.Classification of EvidenceThis study provides Class II evidence that in patients with acute ischemic stroke with CAD risk factors but no known history of CAD, a machine learning model predicts CAD on multidetector coronary CT with an AUC of 0.763 (95% CI 0.711–0.814).

Rationale Very early stage blood pressure (BP) levels may affect outcome in stroke patients who have successfully undergone recanalization following intra-arterial treatment, but the optimal target of BP management remains uncertain. Aim We hypothesized that the clinical outcome after intensive BP-lowering is superior to conventional BP control after successful recanalization by intra-arterial treatment. Sample-size estimates We aim to randomize 668 patients (334 per arm), 1:1. Methods and design We initiated a multicenter, prospective, randomized, open-label trial with a blinded end-point assessment (PROBE) design. After successful recanalization (thrombolysis in cerebral infarction score ≥ 2 b), patients with elevated systolic BP level, defined as the mean of two readings ≥ 140 mmHg, will be randomly assigned to the intensive BP-lowering (systolic BP Study outcomes The primary efficacy outcomes are from dichotomized analysis of modified Rankin Scale (mRS) scores at three months (mRS scores: 0–2 vs. 3–6). The primary safety outcomes are symptomatic intracerebral hemorrhage and death within three months. Discussion The OPTIMAL-BP trial will provide evidence for the effectiveness of active BP control to achieve systolic BP Clinical trial registration ClinicalTrials.gov Identifier: NCT04205305.

Background and Purpose: Patients with acute stroke are often accompanied by comorbidities, such as active cancer. However, adequate treatment guidelines are not available for these patients. The purpose of this study was to evaluate the association between cancer and the outcomes of reperfusion therapy in patients with stroke. Methods: We compared treatment outcomes in patients who underwent reperfusion therapy, using a nationwide reperfusion therapy registry. We divided the patients into 3 groups according to cancer activity: active cancer, nonactive cancer, and without a history of cancer. We investigated reperfusion processes, 24-hour neurological improvement, adverse events, 3-month functional outcome, and 6-month survival and related factors after reperfusion therapy. Results: Among 1338 patients who underwent reperfusion therapy, 62 patients (4.6%) had active cancer, 78 patients (5.8%) had nonactive cancer, and 1198 patients (89.5%) had no history of cancer. Of the enrolled patients, 969 patients received intravenous thrombolysis and 685 patients underwent endovascular treatment (316 patients received combined therapy). Patients with active cancer had more comorbidities and experienced more severe strokes; however, they showed similar 24-hour neurological improvement and adverse events, including cerebral hemorrhage, compared with the other groups. Although the functional outcome at 3 months was poorer than the other groups, 36.4% of patients with active cancer showed functional independence. Additionally, 52.9% of the patients with determined stroke etiology showed functional independence despite active cancer. During the 6-month follow-up, 46.6% of patients with active cancer died, and active cancer was independently associated with poor survival (hazard ratio, 3.973 [95% CI, 2.528–6.245]). Conclusions: In patients with active cancer, reperfusion therapy showed similar adverse events and short-term outcomes to that of other groups. While long-term prognosis was worse in the active cancer group than the nonactive cancer groups, not negligible number of patients had good functional outcomes, especially those with determined stroke mechanisms.

Background and Purpose We aimed to develop a model predicting early recanalization after intravenous tissue plasminogen activator (t-PA) treatment in large-vessel occlusion.Methods Using data from two different multicenter prospective cohorts, we determined the factors associated with early recanalization immediately after t-PA in stroke patients with large-vessel occlusion, and developed and validated a prediction model for early recanalization. Clot volume was semiautomatically measured on thin-section computed tomography using software, and the degree of collaterals was determined using the Tan score. Follow-up angiographic studies were performed immediately after t-PA treatment to assess early recanalization.Results Early recanalization, assessed 61.0±44.7 minutes after t-PA bolus, was achieved in 15.5% (15/97) in the derivation cohort and in 10.5% (8/76) in the validation cohort. Clot volume (odds ratio [OR], 0.979; 95% confidence interval [CI], 0.961 to 0.997; P=0.020) and good collaterals (OR, 6.129; 95% CI, 1.592 to 23.594; P=0.008) were significant factors associated with early recanalization. The area under the curve (AUC) of the model including clot volume was 0.819 (95% CI, 0.720 to 0.917) and 0.842 (95% CI, 0.746 to 0.938) in the derivation and validation cohorts, respectively. The AUC improved when good collaterals were added (derivation cohort: AUC, 0.876; 95% CI, 0.802 to 0.950; P=0.164; validation cohort: AUC, 0.949; 95% CI, 0.886 to 1.000; P=0.036). The integrated discrimination improvement also showed significantly improved prediction (0.097; 95% CI, 0.009 to 0.185; P=0.032).Conclusions The model using clot volume and collaterals predicted early recanalization after intravenous t-PA and had a high performance. This model may aid in determining the recanalization treatment strategy in stroke patients with large-vessel occlusion.

Background Unavailability or saturation of the intensive care unit may be associated with the fatality of COVID-19. Prioritizing the patients for hospitalization and intensive care may be critical for reducing the fatality of COVID-19. This study aimed to develop and validate a new integer-based scoring system for predicting patients with COVID-19 requiring intensive care, using only the predictors available upon triage. Methods This is a retrospective study using cohort data from the Korean Centers for Disease Control and Prevention that included all admitted patients with COVID-19 between January 19 and June 3, 2020, in South Korea. The primary outcome was patients requiring intensive care defined as actual admission to the intensive care unit; at any time use of an extracorporeal life support device, mechanical ventilation, or vasopressors; and death. Patients admitted until March 20 were included for the training dataset to develop the prediction models and externally validated for the patients admitted afterward. Two logistic regression models were developed with different predictors and the predictive performance was compared: one with patient-provided variables and the other with added radiologic and laboratory variables. An integer-based scoring system was developed based on the developed logistic regression model. Results A total of 5193 patients were considered, with 4663 patients included after excluding patients with age under 18 or insufficient data. For the training dataset, 3238 patients were included. Of the included patients, 444 (9.5%) patients required intensive care. The model developed with only the clinical variables showed an area under the curve of 0.884 for the validation set. The performance did not differ when radiologic and laboratory variables were added. Seven variables were selected for developing an integer-based scoring system: age, sex, initial body temperature, dyspnea, hemoptysis, history of chronic kidney disease, and activities of daily living. The area under the curve of the scoring system was 0.880. Conclusions An integer-based scoring system was developed for predicting patients with COVID-19 requiring intensive care, with high performance. This system may aid decision support for prioritizing the patient for hospitalization and intensive care, particularly in a situation with limited medical resources.

We investigated the association of low ankle-brachial index (ABI < 0.9) with major adverse cardiovascular events (MACE) and all-cause mortality in patients with embolic stroke of undetermined source (ESUS) as well as whether the association differed by ESUS subtype. This retrospective single-center study included ESUS patients who underwent transesophageal echocardiography and ABI during hospitalization. ESUS was classified as ESUS with minor cardioembolic source, arteriogenic embolism, two or more causes, or no cause. Arteriogenic embolism was defined and classified as complex aortic or non-stenotic relevant artery plaque. MACE was defined as stroke recurrence, acute coronary syndrome, hospitalization for heart failure, or death. Overall, 829 patients were included, with a median follow-up of 45.8 months. Of these, 42 (5.1%) and 370 (44.6%) had low ABI and arteriogenic embolism, respectively. ABI < 0.9 was independently associated with MACE (hazard ratio [HR]: 2.038, 95% confidence interval [CI]: 1.093–3.801) and all-cause mortality (HR: 3.608, 95% CI: 1.538–8.465) according to the multivariable Cox regression analysis. Between ESUS subtypes, low ABI was independently associated with MACE (HR: 2.513, 95% CI: 1.257–5.023) and all-cause mortality (HR: 5.681, 95% CI: 2.151–15.008) in arteriogenic embolism patients, especially in those with complex aortic plaque. However, in non-arteriogenic embolism patients, low ABI was not related to MACE and mortality. In ESUS patients, low ABI was linked to MACE and all-cause mortality, especially in those with arteriogenic embolisms from complex aortic plaque.

Background: The CHADS2, CHA2DS2-VASc, ATRIA, and Essen scores have been developed for predicting vascular outcomes in stroke patients. We investigated the association between these stroke risk scores and unsuccessful recanalization after endovascular thrombectomy (EVT). Methods: From the nationwide multicenter registry (Selection Criteria in Endovascular Thrombectomy and Thrombolytic therapy (SECRET)) (Clinicaltrials.gov NCT02964052), we consecutively included 501 patients who underwent EVT. We identified pre-admission stroke risk scores in each included patient. Results: Among 501 patients who underwent EVT, 410 (81.8%) patients achieved successful recanalization (mTICI ≥ 2b). Adjusting for body mass index and p < 0.1 in univariable analysis revealed the association between all stroke risk scores and unsuccessful recanalization (CHADS2 score: odds ratio (OR) 1.551, 95% confidence interval (CI) 1.198–2.009, p = 0.001; CHA2DS2VASc score: OR 1.269, 95% CI 1.080–1.492, p = 0.004; ATRIA score: OR 1.089, 95% CI 1.011–1.174, p = 0.024; and Essen score: OR 1.469, 95% CI 1.167–1.849, p = 0.001). The CHADS2 score had the highest AUC value and differed significantly only from the Essen score (AUC of CHADS2 score; 0.618, 95% CI 0.554–0.681). Conclusion: All stroke risk scores were associated with unsuccessful recanalization after EVT. Our study suggests that these stroke risk scores could be used to predict recanalization in stroke patients undergoing EVT.

IntroductionSarcopenia, a age-related disease characterized by loss of muscle mass accompanied by loss of function, is associated with nutrition imbalance, physical inactivity, insulin resistance, inflammation, metabolic syndrome, and atherosclerosis which are risk factors for cardiovascular disease. However, its association with outcomes after ischemic stroke has not been well-established. This study investigated whether functional outcomes of patients with acute ischemic stroke is associated with sarcopenia.MethodsData were collected from 568 consecutive patients with acute ischemic stroke with National Institute of Health Stroke Scale 0–5 or transient ischemic attack who underwent bioelectrical impedance analysis between March 2018 and March 2021. Sarcopenia was defined, as low muscle mass, as measured by bioelectrical impedance analysis, and low muscle strength, as indicated by the Medical Research Council score. Unfavorable functional outcome was defined as mRS score of 2–6 at 90 days after discharge. The relationship between functional outcomes and the presence of sarcopenia or its components was determined.ResultsOf the 568 patients included (mean age 65.5 ± 12.6 years, 64.6% male), sarcopenia was detected in 48 (8.5%). After adjusting for potential confounders, sarcopenia was independently and significantly associated with unfavorable functional outcome (odds ratio 2.37, 95% confidence interval 1.15–4.73 for unfavorable functional outcome, odds ratio 2.10, 95% confidence interval 1.18–3.71 for an increase in the mRS score). Each component of sarcopenia was also independently associated with unfavorable functional outcome (odds ratio 1.76, 95% confidence interval 1.05–2.95 with low muscle mass, odds ratio 2.64, 95% confidence interval 1.64–4.23 with low muscle strength). The impact of low muscle mass was larger in men than in women, and in patients with lower muscle mass of the lower extremities than in those with lower muscle mass of the upper extremities.ConclusionsIn this study, the prevalence of sarcopenia in patients with stroke was lower than most of previous studies and patients with sarcopenia showed higher likelihood for unfavorable functional outcomes at 90 days after acute ischemic stroke or TIA. Further investigation of the interventions for treating sarcopenia and its impact on the outcome of ischemic stroke patients is needed.

Background: The CHADS2, CHA2DS2-VASc, ATRIA, and Essen scores have been developed for predicting vascular outcomes in stroke patients. We investigated the association between these stroke risk scores and unsuccessful recanalization after endovascular thrombectomy (EVT). Methods: From the nationwide multicenter registry (Selection Criteria in Endovascular Thrombectomy and Thrombolytic therapy (SECRET)) (Clinicaltrials.gov NCT02964052), we consecutively included 501 patients who underwent EVT. We identified pre-admission stroke risk scores in each included patient. Results: Among 501 patients who underwent EVT, 410 (81.8%) patients achieved successful recanalization (mTICI ≥ 2b). Adjusting for body mass index and p < 0.1 in univariable analysis revealed the association between all stroke risk scores and unsuccessful recanalization (CHADS2 score: odds ratio (OR) 1.551, 95% confidence interval (CI) 1.198–2.009, p = 0.001; CHA2DS2VASc score: OR 1.269, 95% CI 1.080–1.492, p = 0.004; ATRIA score: OR 1.089, 95% CI 1.011–1.174, p = 0.024; and Essen score: OR 1.469, 95% CI 1.167–1.849, p = 0.001). The CHADS2 score had the highest AUC value and differed significantly only from the Essen score (AUC of CHADS2 score; 0.618, 95% CI 0.554–0.681). Conclusion: All stroke risk scores were associated with unsuccessful recanalization after EVT. Our study suggests that these stroke risk scores could be used to predict recanalization in stroke patients undergoing EVT.

Background: CHADS2, CHA2DS2-VASc, ATRIA, and Essen stroke risk scores are used to estimate thromboembolism risk. We aimed to investigate the association between unfavorable outcomes and stroke risk scores in patients who received endovascular thrombectomy (EVT). Methods: This study was performed using data from a nationwide, multicenter registry to explore the selection criteria for patients who would benefit from reperfusion therapies. We calculated pre-admission CHADS2, CHA2DS2-VASc, ATRIA, and Essen scores for each patient who received EVT and compared the relationship between these scores and 3-month modified Rankin Scale (mRS) records. Results: Among the 404 patients who received EVT, 213 (52.7%) patients had unfavorable outcomes (mRS 3–6). All scores were significantly higher in patients with unfavorable outcomes than in those with favorable outcomes. Multivariable logistic regression analysis indicated that CHADS2 and the ATRIA score were positively correlated with unfavorable outcomes after adjusting for body mass index and variables with p < 0.1 in the univariable analysis (CHADS2 score: odds ratio [OR], 1.484; 95% confidence interval [CI], 1.290–1.950; p = 0.005, ATRIA score, OR, 1.128; 95% CI, 1.041–1.223; p = 0.004). Conclusions: The CHADS2 and ATRIA scores were positively correlated with unfavorable outcomes and could be used to predict unfavorable outcomes in patients who receive EVT.

Additional file 1: Additional Table I. Descriptive statistics of the patients admitted to the intensive care unit. Description of data: A table presenting descriptive statistics of the patients admitted to the intensive care unit. A comparison between the patients who died or required an ECLS device, mechanical ventilator, or vasopressors and the patients who did not is shown.

Background COVID-19 has officially been declared as a pandemic, and the spread of the virus is placing sustained demands on public health systems. There are speculations that the COVID-19 mortality differences between regions are due to the disparities in the availability of medical resources. Therefore, the selection of patients for diagnosis and treatment is essential in this situation. Military personnel are especially at risk for infectious diseases; thus, patient selection with an evidence-based prognostic model is critical for them. Objective This study aims to assess the usability of a novel platform used in the military hospitals in Korea to gather data and deploy patient selection solutions for COVID-19. Methods The platform’s structure was developed to provide users with prediction results and to use the data to enhance the prediction models. Two applications were developed: a patient’s application and a physician’s application. The primary outcome was requiring an oxygen supplement. The outcome prediction model was developed with patients from four centers. A Cox proportional hazards model was developed. The outcome of the model for the patient’s application was the length of time from the date of hospitalization to the date of the first oxygen supplement use. The demographic characteristics, past history, patient symptoms, social history, and body temperature were considered as risk factors. A usability study with the Post-Study System Usability Questionnaire (PSSUQ) was conducted on the physician’s application on 50 physicians. Results The patient’s application and physician’s application were deployed on the web for wider availability. A total of 246 patients from four centers were used to develop the outcome prediction model. A small percentage (n=18, 7.32%) of the patients needed professional care. The variables included in the developed prediction model were age; body temperature; predisease physical status; history of cardiovascular disease; hypertension; visit to a region with an outbreak; and symptoms of chills, feverishness, dyspnea, and lethargy. The overall C statistic was 0.963 (95% CI 0.936-0.99), and the time-dependent area under the receiver operating characteristic curve ranged from 0.976 at day 3 to 0.979 at day 9. The usability of the physician’s application was good, with an overall average of the responses to the PSSUQ being 2.2 (SD 1.1). Conclusions The platform introduced in this study enables evidence-based patient selection in an effortless and timely manner, which is critical in the military. With a well-designed user experience and an accurate prediction model, this platform may help save lives and contain the spread of the novel virus, COVID-19.

BACKGROUND Prioritizing patients in need of intensive care is necessary to reduce the mortality rate during the COVID-19 pandemic. Although several scoring methods have been introduced, many require laboratory or radiographic findings that are not always easily available. OBJECTIVE The purpose of this study was to develop a machine learning model that predicts the need for intensive care for patients with COVID-19 using easily obtainable characteristics—baseline demographics, comorbidities, and symptoms. METHODS A retrospective study was performed using a nationwide cohort in South Korea. Patients admitted to 100 hospitals from January 25, 2020, to June 3, 2020, were included. Patient information was collected retrospectively by the attending physicians in each hospital and uploaded to an online case report form. Variables that could be easily provided were extracted. The variables were age, sex, smoking history, body temperature, comorbidities, activities of daily living, and symptoms. The primary outcome was the need for intensive care, defined as admission to the intensive care unit, use of extracorporeal life support, mechanical ventilation, vasopressors, or death within 30 days of hospitalization. Patients admitted until March 20, 2020, were included in the derivation group to develop prediction models using an automated machine learning technique. The models were externally validated in patients admitted after March 21, 2020. The machine learning model with the best discrimination performance was selected and compared against the CURB-65 (confusion, urea, respiratory rate, blood pressure, and 65 years of age or older) score using the area under the receiver operating characteristic curve (AUC). RESULTS A total of 4787 patients were included in the analysis, of which 3294 were assigned to the derivation group and 1493 to the validation group. Among the 4787 patients, 460 (9.6%) patients needed intensive care. Of the 55 machine learning models developed, the XGBoost model revealed the highest discrimination performance. The AUC of the XGBoost model was 0.897 (95% CI 0.877-0.917) for the derivation group and 0.885 (95% CI 0.855-0.915) for the validation group. Both the AUCs were superior to those of CURB-65, which were 0.836 (95% CI 0.825-0.847) and 0.843 (95% CI 0.829-0.857), respectively. CONCLUSIONS We developed a machine learning model comprising simple patient-provided characteristics, which can efficiently predict the need for intensive care among patients with COVID-19.

BACKGROUND Clear guidelines for a patient with suspected COVID-19 infection are unavailable. Many countries rely on assessments through a national hotline or telecommunications, but this only adds to the burden of an already overwhelmed health care system. In this study, we developed an algorithm and a web application to help patients get screened. OBJECTIVE This study aims to aid the general public by developing a web-based application that helps patients decide when to seek medical care during a novel disease outbreak. METHODS The algorithm was developed via consultations with 6 physicians who directly screened, diagnosed, and/or treated patients with COVID-19. The algorithm mainly focused on when to test a patient in order to allocate limited resources more efficiently. The application was designed to be mobile-friendly and deployed on the web. We collected the application usage pattern data from March 1 to March 27, 2020. We evaluated the association between the usage pattern and the numbers of COVID-19 confirmed, screened, and mortality cases by access location and digital literacy by age group. RESULTS The algorithm used epidemiological factors, presence of fever, and other symptoms. In total, 83,460 users accessed the application 105,508 times. Despite the lack of advertisement, almost half of the users accessed the application from outside of Korea. Even though the digital literacy of the 60+ years age group is half of that of individuals in their 50s, the number of users in both groups was similar for our application. CONCLUSIONS We developed an expert-opinion–based algorithm and web-based application for screening patients. This innovation can be helpful in circumstances where information on a novel disease is insufficient and may facilitate efficient medical resource allocation.

We investigated whether there was an annual change in outcomes in patients who received the thrombolytic therapy or endovascular treatment (EVT) in Korea. This analysis was performed using data from a nationwide multicenter registry for exploring the selection criteria of patients who would benefit from reperfusion therapies in Korea. We compared the annual changes in the modified Rankin scale (mRS) at discharge and after 90 days and the achievement of successful recanalization from 2012 to 2017. We also investigated the determinants of favorable functional outcomes. Among 1230 included patients, the improvement of functional outcome at discharge after reperfusion therapy was noted as the calendar year increased (p &lt, 0.001). The proportion of patients who were discharged to home significantly increased (from 45.6% in 2012 to 58.5% in 2017) (p &lt, 0.001). The successful recanalization rate increased over time from 78.6% in 2012 to 85.1% in 2017 (p = 0.006). Time from door to initiation of reperfusion therapy decreased over the years (p &lt, 0.05). These secular trends of improvements were also observed in 1203 patients with available mRS data at 90 days (p &lt, 0.05). Functional outcome was associated with the calendar year, age, initial stroke severity, diabetes, preadmission disability, intervals from door to reperfusion therapy, and achievement of successful recanalization. This study demonstrated the secular trends of improvement in functional outcome and successful recanalization rate in patients who received reperfusion therapy in Korea.

We previously developed learning models for predicting the need for intensive care and oxygen among patients with coronavirus disease (COVID-19). Here, we aimed to prospectively validate the accuracy of these models.Probabilities of the need for intensive care [intensive care unit (ICU) score] and oxygen (oxygen score) were calculated from information provided by hospitalized COVID-19 patients (n=44) via a web-based application. The performance of baseline scores to predict 30-day outcomes was assessed.Among 44 patients, 5 and 15 patients needed intensive care and oxygen, respectively. The area under the curve of ICU score and oxygen score to predict 30-day outcomes were 0.774 [95% confidence interval (CI): 0.614-0.934] and 0.728 (95% CI: 0.559-0.898), respectively. The ICU scores of patients needing intensive care increased daily by 0.71 points (95% CI: 0.20-1.22) after hospitalization and by 0.85 points (95% CI: 0.36-1.35) after symptom onset, which were significantly different from those in individuals not needing intensive care (Our machine learning models showed good performance for predicting the outcomes of COVID-19 patients and could thus be useful for patient triage and monitoring.

Background and purpose We aimed to determine whether the care process and outcomes in patients with acute stroke who received recanalization therapy changed during the outbreak of coronavirus disease 2019 (COVID-19) in South Korea. Methods We used data from a prospective multicenter reperfusion therapy registry to compare the care process-including the time from symptom onset to treatment, number of treated patients, and discharge disposition-and treatment outcomes between before and during the COVID-19 outbreak in South Korea. Results Upon the COVID-19 outbreak in South Korea, the number of patients receiving endovascular treatment to decrease temporarily but considerably. The use of emergency medical services by stroke patients increased from 91.5% before to 100.0% during the COVID-19 outbreak (p=0.025), as did the median time from symptom onset to hospital visit [median (interquartile range), 91.0 minutes (39.8-277.0) vs. 176.0 minutes (56.0-391.5), p=0.029]. Furthermore, more functionally dependent patients with disabilities were discharged home (59.5% vs. 26.1%, p=0.020) rather than staying in a regional or rehabilitation hospital. In contrast, there were no COVID-19-related changes in the times from the hospital visit to brain imaging and treatment or in the functional outcome, successful recanalization rate, or rate of symptomatic intracerebral hemorrhage. Conclusions These findings suggest that a prehospital delay occurred during the COVID-19 outbreak, and that patients with acute stroke might have been reluctant to visit and stay in hospitals. Our findings indicate that attention should be paid to prehospital care and the behavior of patients with acute stroke during the COVID-19 outbreak.

Background Prioritizing patients in need of intensive care is necessary to reduce the mortality rate during the COVID-19 pandemic. Although several scoring methods have been introduced, many require laboratory or radiographic findings that are not always easily available. Objective The purpose of this study was to develop a machine learning model that predicts the need for intensive care for patients with COVID-19 using easily obtainable characteristics—baseline demographics, comorbidities, and symptoms. Methods A retrospective study was performed using a nationwide cohort in South Korea. Patients admitted to 100 hospitals from January 25, 2020, to June 3, 2020, were included. Patient information was collected retrospectively by the attending physicians in each hospital and uploaded to an online case report form. Variables that could be easily provided were extracted. The variables were age, sex, smoking history, body temperature, comorbidities, activities of daily living, and symptoms. The primary outcome was the need for intensive care, defined as admission to the intensive care unit, use of extracorporeal life support, mechanical ventilation, vasopressors, or death within 30 days of hospitalization. Patients admitted until March 20, 2020, were included in the derivation group to develop prediction models using an automated machine learning technique. The models were externally validated in patients admitted after March 21, 2020. The machine learning model with the best discrimination performance was selected and compared against the CURB-65 (confusion, urea, respiratory rate, blood pressure, and 65 years of age or older) score using the area under the receiver operating characteristic curve (AUC). Results A total of 4787 patients were included in the analysis, of which 3294 were assigned to the derivation group and 1493 to the validation group. Among the 4787 patients, 460 (9.6%) patients needed intensive care. Of the 55 machine learning models developed, the XGBoost model revealed the highest discrimination performance. The AUC of the XGBoost model was 0.897 (95% CI 0.877-0.917) for the derivation group and 0.885 (95% CI 0.855-0.915) for the validation group. Both the AUCs were superior to those of CURB-65, which were 0.836 (95% CI 0.825-0.847) and 0.843 (95% CI 0.829-0.857), respectively. Conclusions We developed a machine learning model comprising simple patient-provided characteristics, which can efficiently predict the need for intensive care among patients with COVID-19.

Introduction: Prediction of outcome in stroke patients can help both physicians and patients in making treatment decisions and managing prognostic expectations. Machine learning techniques are being increasingly used in the field of medical research. Hypothesis: We hypothesized that models developed with machine learning techniques are useful for predicting long-term functional outcomes in patients with acute ischemic stroke. Methods: The model was developed with a prospective cohort of acute ischemic stroke patients. This cohort registers stroke patients who were admitted within 7 days from the onset of symptoms. For this study, we included all patients admitted between January 1, 2010, and December 31, 2014. We excluded patients with pre-stroke modified Rankin Scale (mRS) score >2 or missing 3-month mRS score. Univariate analysis was performed to guide in variable selection for machine learning models. Machine learning models were trained to classify patients likely to have unfavorable outcome, defined as 3-month mRS score >2. Developed models included artificial neural network, random forest, support vector machine, and logistic regression models. Area under the receiver-operated characteristic curve (AUC) was used to compare effectiveness of each model. Google’s TensorFlow and scikit-learn toolkit were used for training of machine learning models. Results: Total of 3,524 patients were admitted during the study period. After excluding 454 patients with unavailable 3-month mRS scores, 60 patients with pre-stroke mRS score >2, and 87 patients with missing laboratory tests or clinical data, 2,923 patients were finally enrolled. Of the 2,923 patients, 695 (24%) patients had unfavorable outcome (mRS>2) at 3 months. The AUC was 0.888 for Artificial neural network model, 0.810 for random forest model, 0.836 for support vector machine model, and 0.842 for logistic regression model. We calculated the AUC of ASTRAL score for reference, which showed 0.839 in our study group. Conclusion: Machine learning models, particularly the artificial neural network model, achieved high accuracy of prediction for functional outcome in stroke. This study showed the feasibility of machine learning approach for predicting outcomes in stroke.

The major challenge in the diagnosis of disseminated intravascular coagulation (DIC) comes from the lack of specific biomarkers, leading to developing composite scoring systems. DIC scores are simple and rapidly applicable. However, optimal fibrin-related markers and their cut-off values remain to be defined, requiring optimization for use. The aim of this study is to optimize the use of DIC-related parameters through machine learning (ML)-approach. Further, we evaluated whether this approach could provide a diagnostic value in DIC diagnosis. For this, 46 DIC-related parameters were investigated for both clinical findings and laboratory results. We retrospectively reviewed 656 DIC-suspected cases at an initial order for full DIC profile and labeled their evaluation results (Set 1; DIC, n = 228; non-DIC, n = 428). Several ML algorithms were tested, and an artificial neural network (ANN) model was established via independent training and testing using 32 selected parameters. This model was externally validated from a different hospital with 217 DIC-suspected cases (Set 2; DIC, n = 80; non-DIC, n = 137). The ANN model represented higher AUC values than the three scoring systems in both set 1 (ANN 0.981; ISTH 0.945; JMHW 0.943; and JAAM 0.928) and set 2 (AUC ANN 0.968; ISTH 0.946). Additionally, the relative importance of the 32 parameters was evaluated. Most parameters had contextual importance, however, their importance in ML-approach was different from the traditional scoring system. Our study demonstrates that ML could optimize the use of clinical parameters with robustness for DIC diagnosis. We believe that this approach could play a supportive role in physicians' medical decision by integrated into electrical health record system. Further prospective validation is required to assess the clinical consequence of ML-approach and their clinical benefit.

Introduction: The presence of spontaneous echo contrast (SEC) on transesophageal echocardiography (TEE) is associated with increased thrombogenecity and a risk factor of thromboembolism. The increased thrombogenecity in SEC may produce larger intracardiac thrombi, which result in larger cerebral infarctions and severe stroke. Hypothesis: We hypothesized that infarction volume will be larger in stroke patients with SEC than in those without SEC. Methods: This was a post-hoc analysis using a prospective cohort of acute ischemic stroke. This study included patients with nonvalvular atrial fibrillation (NVAF) who underwent TEE and diffusion weighted imaging (DWI) from Jan. 2008 to Dec. 2014. The volume of cerebral infarction on DWI was measured semi-automatically using 3-dimensional software by an investigator who was blinded to clinical information. The infarction volume was compared between patients with SEC and those without. Results: Of 4252 considered patients, 889 patients had NVAF. After excluding 449 patients without TEE and 39 patients without DWI, 401 patients were included for analysis. Of them, SEC was found in 181 patients (45.1%, 82 mild, 35 moderate, and 64 severe SEC). Infarction volume was larger for the patients with SEC than those without SEC (median [interquartile range], 7226.8 mm 3 [1218.1-28804.6] vs. 4756.8 mm 3 [672.8-14887.8], p=0.015). Infarction volume also increased with SEC severity (p=0.006). Initial National Institute of Health Stroke Scale scores were higher for the patients with SEC than those without SEC (median [interquartile range], 5.0 [2.0-12.0] vs. 3.0 [1.0-8.0]. p=0.008). On multivariate analysis, infarction volume was independently associated with the presence of SEC (p=0.014). Conclusion: Among stroke patients with NVAF, those with co-existing SEC had larger cerebral infarction, which may account for severe stroke. This may be related with increased thrombogenecity in patients with SEC.