ABSTRACT
OBJECTIVE: To develop and externally validate models based on neural networks and natural language processing (NLP) to identify suspected serious infections in emergency department (ED) patients afebrile at initial presentation. METHODS: This retrospective study included adults who visited the ED afebrile at initial presentation. We developed four models based on artificial neural networks to identify suspected serious infection. Patient demographics, vital signs, laboratory test results and information extracted from initial ED physician notes using term frequency-inverse document frequency were used as model variables. Models were trained and internally validated with data from one hospital and externally validated using data from a different hospital. Model discrimination was evaluated using area under the receiver operating characteristic curve (AUC) and 95% confidence intervals (CIs). RESULTS: The training, internal validation, and external validation datasets comprised 150,699, 37,675, and 85,098 patients, respectively. The AUCs (95% CIs) for Models 1 (demographics + vital signs), 2 (demographics + vital signs + initial ED physician note), 3 (demographics + vital signs + laboratory tests), and 4 (demographics + vital signs + laboratory tests + initial ED physician note) in the internal validation dataset were 0.789 (0.782-0.796), 0.867 (0.862-0.872), 0.881 (0.876-0.887), and 0.911 (0.906-0.915), respectively. In the external validation dataset, the AUCs (95% CIs) of Models 1, 2, 3, and 4 were 0.824 (0.817-0.830), 0.895 (0.890-0.899), 0.879 (0.873-0.884), and 0.913 (0.909-0.917), respectively. Model 1 can be utilized immediately after ED triage, Model 2 can be utilized after the initial physician notes are recorded (median time from ED triage: 28 min), and Models 3 and 4 can be utilized after the initial laboratory tests are reported (median time from ED triage: 68 min). CONCLUSIONS: We developed and validated models to identify suspected serious infection in the ED. Extracted information from initial ED physician notes using NLP contributed to increased model performance, permitting identification of suspected serious infection at early stages of ED visits.
Subject(s)
Emergency Service, Hospital , Natural Language Processing , Neural Networks, Computer , Humans , Retrospective Studies , Female , Male , Middle Aged , Adult , Early Diagnosis , Aged , ROC Curve , Infections/diagnosis , Vital Signs , Electronic Health RecordsABSTRACT
OBJECTIVES: In the era of the Fourth Industrial Revolution, where an ecosystem is being developed to enhance the quality of healthcare services by applying information and communication technologies, systematic and sustainable data management is essential for medical institutions. In this study, we assessed the data management status and emerging concerns of three medical institutions, while also examining future directions for seamless data management. METHODS: To evaluate the data management status, we examined data types, capacities, infrastructure, backup methods, and related organizations. We also discussed challenges, such as resource and infrastructure issues, problems related to government regulations, and considerations for future data management. RESULTS: Hospitals are grappling with the increasing data storage space and a shortage of management personnel due to costs and project termination, which necessitates countermeasures and support. Data management regulations on the destruction or maintenance of medical records are needed, and institutional consideration for secondary utilization such as long-term treatment or research is required. Government-level guidelines for facilitating hospital data sharing and mobile patient services should be developed. Additionally, hospital executives at the organizational level need to make efforts to facilitate the clinical validation of artificial intelligence software. CONCLUSIONS: This analysis of the current status and emerging issues of data management reveals potential solutions and sets the stage for future organizational and policy directions. If medical big data is systematically managed, accumulated over time, and strategically monetized, it has the potential to create new value.
ABSTRACT
BACKGROUND: A clinical decision support system (CDSS) is recognized as a technology that enhances clinical efficacy and safety. However, its full potential has not been realized, mainly due to clinical data standards and noninteroperable platforms. OBJECTIVE: In this paper, we introduce the common data model-based intelligent algorithm network environment (CANE) platform that supports the implementation and deployment of a CDSS. METHODS: CDSS reasoning engines, usually represented as R or Python objects, are deployed into the CANE platform and converted into C# objects. When a clinician requests CANE-based decision support in the electronic health record (EHR) system, patients' information is transformed into Health Level 7 Fast Healthcare Interoperability Resources (FHIR) format and transmitted to the CANE server inside the hospital firewall. Upon receiving the necessary data, the CANE system's modules perform the following tasks: (1) the preprocessing module converts the FHIRs into the input data required by the specific reasoning engine, (2) the reasoning engine module operates the target algorithms, (3) the integration module communicates with the other institutions' CANE systems to request and transmit a summary report to aid in decision support, and (4) creates a user interface by integrating the summary report and the results calculated by the reasoning engine. RESULTS: We developed a CANE system such that any algorithm implemented in the system can be directly called through the RESTful application programming interface when it is integrated with an EHR system. Eight algorithms were developed and deployed in the CANE system. Using a knowledge-based algorithm, physicians can screen patients who are prone to sepsis and obtain treatment guides for patients with sepsis with the CANE system. Further, using a nonknowledge-based algorithm, the CANE system supports emergency physicians' clinical decisions about optimum resource allocation by predicting a patient's acuity and prognosis during triage. CONCLUSIONS: We successfully developed a common data model-based platform that adheres to medical informatics standards and could aid artificial intelligence model deployment using R or Python.
Subject(s)
Decision Support Systems, Clinical , Sepsis , Artificial Intelligence , Electronic Health Records , Health Level Seven , Humans , Knowledge BasesABSTRACT
BACKGROUND: Low-density lipoprotein-cholesterol (LDL-C) is used as a threshold and target for treating dyslipidemia. Although the Friedewald equation is widely used to estimate LDL-C, it has been known to be inaccurate in the case of high triglycerides (TG) or non-fasting states. We aimed to propose a novel method to estimate LDL-C using machine learning. METHODS: Using a large, single-center electronic health record database, we derived a ML algorithm to estimate LDL-C from standard lipid profiles. From 1,029,572 cases with both standard lipid profiles (total cholesterol, high-density lipoprotein-cholesterol, and TG) and direct LDL-C measurements, 823,657 tests were used to derive LDL-C estimation models. Patient characteristics such as sex, age, height, weight, and other laboratory values were additionally used to create separate data sets and algorithms. RESULTS: Machine learning with gradient boosting (LDL-CX) and neural network (LDL-CN) showed better correlation with directly measured LDL-C, compared with conventional methods (r = 0.9662, 0.9668, 0.9563, 0.9585; for LDL-CX, LDL-CN, Friedewald [LDL-CF], and Martin [LDL-CM] equations, respectively). The overall bias of LDL-CX (-0.27 mg/dL, 95% CI -0.30 to -0.23) and LDL-CN (-0.01 mg/dL, 95% CI -0.04-0.03) were significantly smaller compared with both LDL-CF (-3.80 mg/dL, 95% CI -3.80 to -3.60) or LDL-CM (-2.00 mg/dL, 95% CI -2.00 to -1.94), especially at high TG levels. CONCLUSIONS: Machine learning algorithms were superior in estimating LDL-C compared with the conventional Friedewald or the more contemporary Martin equations. Through external validation and modification, machine learning could be incorporated into electronic health records to substitute LDL-C estimation.
Subject(s)
Cholesterol, LDL/analysis , Dyslipidemias/diagnosis , Machine Learning , Algorithms , Cholesterol, HDL , Humans , TriglyceridesABSTRACT
BACKGROUND: Digital health care is an important strategy in the war against COVID-19. South Korea introduced living and treatment support centers (LTSCs) to control regional outbreaks and care for patients with asymptomatic or mild COVID-19. Seoul National University Hospital (SNUH) introduced information and communications technology (ICT)-based solutions to manage clinically healthy patients with COVID-19. OBJECTIVE: This study aims to investigate satisfaction and usability by patients and health professionals in the optimal use of a mobile app and wearable device that SNUH introduced to the LTSC for clinically healthy patients with COVID-19. METHODS: Online surveys and focus group interviews were conducted to collect quantitative and qualitative data. RESULTS: Regarding usability testing of the wearable device, perceived usefulness had the highest mean score of 4.45 (SD 0.57) points out of 5. Regarding usability of the mobile app, perceived usefulness had the highest mean score of 4.62 (SD 0.48) points out of 5. Regarding satisfaction items for the mobile app among medical professionals, the "self-reporting" item had the highest mean score of 4.42 (SD 0.58) points out of 5. In focus group interviews of health care professionals, hospital information system interfacing was the most important functional requirement for ICT-based COVID-19 telemedicine. CONCLUSIONS: Improvement of patient safety and reduction of the burden on medical staff were the expected positive outcomes. Stability and reliability of the device, patient education, accountability, and reimbursement issues should be considered as part of the development of remote patient monitoring. In responding to a novel contagious disease, telemedicine and a wearable device were shown to be useful during a global crisis.
Subject(s)
Asthma/complications , COVID-19/virology , SARS-CoV-2 , Virus Shedding , Adult , Age Factors , Aged , COVID-19/mortality , Female , Humans , Male , Middle Aged , Risk Factors , SARS-CoV-2/isolation & purification , Time Factors , Young AdultABSTRACT
BACKGROUND: South Korea took preemptive action against coronavirus disease (COVID-19) by implementing extensive testing, thorough epidemiological investigation, strict social distancing, and rapid treatment of patients according to disease severity. The Korean government entrusted large-scale hospitals with the operation of living and treatment support centers (LTSCs) for the management for clinically healthy COVID-19 patients. OBJECTIVE: The aim of this paper is to introduce our experience implementing information and communications technology (ICT)-based remote patient management systems at a COVID-19 LTSC. METHODS: We adopted new electronic health record templates, hospital information system (HIS) dashboards, cloud-based medical image sharing, a mobile app, and smart vital sign monitoring devices. RESULTS: Enhancements were made to the HIS to assist in the workflow and care of patients in the LTSC. A dashboard was created for the medical staff to view the vital signs and symptoms of all patients. Patients used a mobile app to consult with their physician or nurse, answer questionnaires, and input self-measured vital signs; the results were uploaded to the hospital information system in real time. Cloud-based image sharing enabled interoperability between medical institutions. Korea's strategy of aggressive mitigation has "flattened the curve" of the rate of infection. A multidisciplinary approach was integral to develop systems supporting patient care and management at the living and treatment support center as quickly as possible. CONCLUSIONS: Faced with a novel infectious disease, we describe the implementation and experience of applying an ICT-based patient management system in the LTSC affiliated with Seoul National University Hospital. ICT-based tools and applications are increasingly important in health care, and we hope that our experience will provide insight into future technology-based infectious disease responses.
Subject(s)
Coronavirus Infections/diagnosis , Coronavirus Infections/therapy , Hospitals, University/organization & administration , Information Technology , Pneumonia, Viral/diagnosis , Pneumonia, Viral/therapy , Adult , Asymptomatic Diseases/epidemiology , Betacoronavirus/isolation & purification , COVID-19 , COVID-19 Testing , Clinical Laboratory Techniques , Coronavirus Infections/drug therapy , Coronavirus Infections/virology , Electronic Health Records , Female , Humans , Male , Mobile Applications , Pandemics , Pneumonia, Viral/virology , Republic of Korea/epidemiology , SARS-CoV-2 , Telemedicine , COVID-19 Drug TreatmentSubject(s)
Anti-Bacterial Agents/pharmacology , Asthma/microbiology , Azithromycin/pharmacology , Cefixime/pharmacology , Gastrointestinal Microbiome/drug effects , Allergens/immunology , Animals , Asthma/immunology , Asthma/therapy , Bronchoalveolar Lavage Fluid/cytology , Bronchoalveolar Lavage Fluid/immunology , Bronchoalveolar Lavage Fluid/microbiology , Cytokines/immunology , Fecal Microbiota Transplantation , Feces/microbiology , Female , Mice, Inbred BALB C , Ovalbumin/immunology , Th2 Cells/drug effects , Th2 Cells/immunologyABSTRACT
OBJECTIVES: Triage is a process to accurately assess and classify symptoms to identify and provide rapid treatment to patients. The Korean Triage and Acuity Scale (KTAS) is used as a triage instrument in all emergency centers. The aim of this study was to train and compare machine learning models to predict KTAS levels. METHODS: This was a cross-sectional study using data from a single emergency department of a tertiary university hospital. Information collected during triage was used in the analysis. Logistic regression, random forest, and XGBoost were used to predict the KTAS level. RESULTS: The models with the highest area under the receiver operating characteristic curve (AUROC) were the random forest and XGBoost models trained on the entire dataset (AUROC = 0.922, 95% confidence interval 0.917-0.925 and AUROC = 0.922, 95% confidence interval 0.918-0.925, respectively). The AUROC of the models trained on the clinical data was higher than that of models trained on text data only, but the models trained on all variables had the highest AUROC among similar machine learning models. CONCLUSIONS: Machine learning can robustly predict the KTAS level at triage, which may have many possibilities for use, and the addition of text data improves the predictive performance compared to that achieved by using structured data alone.
ABSTRACT
Although the symptoms of food allergy are diverse and sometimes dangerous to life, we have not been able to effectively share information on patients' food allergies. In this study, we developed a program to prescribe the allergic formula for each patient and to establish a series of processes for safe allergenic meal delivery based on the standards and guidelines of food allergy management. We then assessed the utility of the "introduction of food allergy program" by comparing the number of allergic prescriptions before and after the program application for inpatients. Through the development and introduction of the program, all hospital staffs, including medical staff and dieticians, can share information on food allergy patients. Systematic management of food allergy patients from doctor's prescriptions has provided the basis for safe meal preparation.
Subject(s)
Food Hypersensitivity , Allergens , Humans , Information Dissemination , Nutritionists , PrescriptionsABSTRACT
Malicious e-mails sent intentionally to institutions have caused an increase in data breaches. Measures against these methods must be taken by healthcare institutions to prevent leakage of sensitive personal medical information. As a form of training, we conducted a phishing simulation to gauge the response of the hospital staff to similar email attacks, and to raise awareness about information security protocols.
Subject(s)
Electronic Mail , HospitalsABSTRACT
BACKGROUND: This study aimed to train, validate and compare predictive models that use machine learning analysis for good neurological recovery in OHCA patients. METHODS: Adult OHCA patients who had a presumed cardiac etiology and a sustained return of spontaneous circulation between 2013 and 2016 were analyzed; 80% of the individuals were analyzed for training and 20% were analyzed for validation. We developed using six machine learning algorithms: logistic regression (LR), extreme gradient boosting (XGB), support vector machine, random forest, elastic net (EN), and neural network. Variables that could be obtained within 24â¯hours of the emergency department visit were used. The area under the receiver operation curve (AUROC) was calculated to assess the discrimination. Calibration was assessed by the Hosmer-Lemeshow test. Reclassification was assessed by using the continuous net reclassification index (NRI). RESULTS: A total of 19,860 OHCA patients were included in the analysis. Of the 15,888 patients in the training group, 2228 (14.0%) had a good neurological recovery; of the 3972 patients in the validation group, 577 (14.5%) had a good neurological recovery. The LR, XGB, and EN models showed the highest discrimination powers (AUROC (95% CI)) of 0.949 (0.941-0.957) for all), and all three models were well calibrated (Hosmer-Lemeshow test: p >0.05). The XGB model reclassified patients according to their true risk better than the LR model (NRI: 0.110), but the EN model reclassified patients worse than the LR model (NRI: -1.239). CONCLUSION: The best performing machine learning algorithm was the XGB and LR algorithm.
Subject(s)
Cardiopulmonary Resuscitation , Clinical Decision Rules , Machine Learning , Nervous System Diseases , Out-of-Hospital Cardiac Arrest , Cardiopulmonary Resuscitation/adverse effects , Cardiopulmonary Resuscitation/methods , Heart Diseases/complications , Humans , Nervous System Diseases/diagnosis , Nervous System Diseases/etiology , Out-of-Hospital Cardiac Arrest/complications , Out-of-Hospital Cardiac Arrest/etiology , Out-of-Hospital Cardiac Arrest/therapy , Prognosis , Recovery of Function , Reproducibility of ResultsABSTRACT
BACKGROUND: Despite the advances in the diagnosis and treatment of heart failure (HF), the current hospital-oriented framework for HF management does not appear to be sufficient to maintain the stability of HF patients in the long term. The importance of self-care management is increasingly being emphasized as a promising long-term treatment strategy for patients with chronic HF. OBJECTIVE: The objective of this study was to evaluate whether a new information communication technology (ICT)-based telehealth program with voice recognition technology could improve clinical or laboratory outcomes in HF patients. METHODS: In this prospective single-arm pilot study, we recruited 31 consecutive patients with chronic HF who were referred to our institute. An ICT-based telehealth program with voice recognition technology was developed and used by patients with HF for 12 weeks. Patients were educated on the use of this program via mobile phone, landline, or the Internet for the purpose of improving communication and data collection. Using these systems, we collected comprehensive data elements related to the risk of HF self-care management such as weight, diet, exercise, medication adherence, overall symptom change, and home blood pressure. The study endpoints were the changes observed in urine sodium concentration (uNa), Minnesota Living with Heart Failure (MLHFQ) scores, 6-min walk test, and N-terminal prohormone of brain natriuretic peptide (NT-proBNP) as surrogate markers for appropriate HF management. RESULTS: Among the 31 enrolled patients, 27 (87%) patients completed the study, and 10 (10/27, 37%) showed good adherence to ICT-based telehealth program with voice recognition technology, which was defined as the use of the program for 100 times or more during the study period. Nearly three-fourths of the patients had been hospitalized at least once because of HF before the enrollment (20/27, 74%); 14 patients had 1, 2 patients had 2, and 4 patients had 3 or more previous HF hospitalizations. In the total study population, there was no significant interval change in laboratory and functional outcome variables after 12 weeks of ICT-based telehealth program. In patients with good adherence to ICT-based telehealth program, there was a significant improvement in the mean uNa (103.1 to 78.1; P=.01) but not in those without (85.4 to 96.9; P=.49). Similarly, a marginal improvement in MLHFQ scores was only observed in patients with good adherence (27.5 to 21.4; P=.08) but not in their counterparts (19.0 to 19.7; P=.73). The mean 6-min walk distance and NT-proBNP were not significantly increased in patients regardless of their adherence. CONCLUSIONS: Short-term application of ICT-based telehealth program with voice recognition technology showed the potential to improve uNa values and MLHFQ scores in HF patients, suggesting that better control of sodium intake and greater quality of life can be achieved by this program.
ABSTRACT
BACKGROUND: The use of mild therapeutic hypothermia (TH) in out-of-hospital cardiac arrest (OHCA) with shockable rhythms is recommended and widely used. However, it is unclear whether TH is associated with better outcomes in non-shockable rhythms. METHODS: This is a retrospective observational study using a national OHCA cohort database composed of emergency medical services (EMS) and hospital data. We included adult EMS-treated OHCA patients of presumed cardiac etiology who were admitted to the hospital during Jan. 2008 to Dec. 2013. Patients without hospital outcome data were excluded. The primary outcome was good neurological outcome at discharge; secondary outcome was survival to discharge. The primary exposure was TH. We compared outcomes between TH and non-TH groups using multivariable logistic regression, adjusting for individual and Utstein factors. Interactions of initial ECG rhythm and witnessed status on the effect of TH on outcomes were tested. RESULTS: There were 11,256 patients in the final analysis. TH was performed in 1703 patients (15.1%). Neurological outcome was better in TH (23.5%) than non-TH (15.0%) (Adjusted OR=1.25, 95% CI 1.05-1.48). The effect of TH on the odds for good neurological outcome was highest in the witnessed PEA group (Adjusted OR=3.91, 95% CI 1.87-8.14). Survival to discharge was significantly higher in the TH group (55.1%) than non-TH (35.9%) (Adjusted OR=1.76, 95% CI 1.56-2.00). CONCLUSIONS: In a nationwide observational study, TH is associated with better neurological outcome and higher survival to discharge. The effect of TH is greatest in witnessed OHCA patients with PEA as the initial ECG rhythm.
Subject(s)
Electrocardiography/methods , Hypothermia, Induced/methods , Out-of-Hospital Cardiac Arrest/therapy , Adult , Aged , Aged, 80 and over , Cardiopulmonary Resuscitation , Databases, Factual , Emergency Medical Services , Female , Hospitals , Humans , Logistic Models , Male , Middle Aged , Patient Discharge , Registries , Republic of Korea , Retrospective Studies , Survival Rate , Treatment OutcomeABSTRACT
BACKGROUND: The 2013 ACCF/AHA guideline for the management of ST elevation myocardial infarction (STEMI) recommends that patients be transported by emergency medical services (EMS) directly to a percutaneous coronary intervention (PCI)-capable hospital. We examined the effects of EMS use according to inter-hospital transfer on time to PCI in STEMI patients. METHODS: Adult patients diagnosed with STEMI from November 2007 to December 2012 with symptom onset less than 24 hours treated with primary PCI at 29 emergency departments (ED) were included. Patients with unknown information about important time variables, inter-hospital transfer and EMS use, and patients who already received PCI at another hospital were excluded. Patients were divided into groups according to EMS use and inter-hospital transfer: Group A (direct to final ED by EMS), Group B (transferred to final ED after EMS transport), Group C (direct to final ED not by EMS), and Group D (transferred to final ED after non-EMS transport). Symptom to balloon time less than 120 minutes was considered timely PCI. Multivariable logistic regression model adjusting for potential risk factors examined the relationship between the groups and timely PCI. Interactions between EMS use and inter-hospital transfer were also tested for the outcome. RESULTS: A total of 5826 patients were analyzed in this study, of which 28.3% called for EMS and 50% were transferred to another hospital for PCI. Median symptom to balloon time was 216 minutes. Timely PCI was achieved in 20.3% of the patients. With the Group D as the reference, the adjusted odds ratio (AOR) with 95% confidence intervals (95% CI) for timely PCI was 5.78 (4.81-6.95) for Group A, 0.80 (0.53-1.20) for Group B, and 2.87 (2.39-3.44) for Group C. In the interaction model, the AOR (95% CI) of EMS use in nontransferred groups and transferred groups was 2.01(1.71-2.38) and 0.80(0.53-1.20). CONCLUSIONS: EMS use significantly increased the odds of timely primary PCI to patients directly transported to a primary PCI center, but not in patients transferred from another hospital. EMS systems that identify STEMI patients and transport them to PCI capable hospitals, and processes to expedite the transfer of patients between non-PCI and PCI hospitals need to be developed further.
Subject(s)
Emergency Medical Services/statistics & numerical data , Myocardial Infarction/therapy , Patient Transfer/statistics & numerical data , Percutaneous Coronary Intervention , Time-to-Treatment , Aged , Electrocardiography , Female , Humans , Male , Middle Aged , Practice Guidelines as Topic , Republic of KoreaABSTRACT
The aim the study was to determine the effect of SonoPrep® on the delivery and analgesic effects of EMLA® cream prior to intravenous (iv) cannulation in a tertiary pediatric emergency department. Children aged between 5 and 10 years were enrolled. Patients were randomized to receive either sonophoresis with SonoPrep® or sham sonophoresis followed by application of EMLA® cream for 5 min prior to iv cannulation. The primary outcome measurement was the child's rating of pain immediately after iv placement, using a 10-cm visual analog scale (VAS). Parents or guardians and blinded researchers were additionally asked to rate their perception of the child's pain using the 10-cm VAS and the Wong-Baker Face scale. A total of 42 patients completed the study (21 in the study group, 21 in the control group). The baseline characteristics between the groups were similar. The VAS pain score was significantly lower in children treated with sonophoresis compared with the sham sonophoresis (median (percentiles 25th-75th), 20.0 (10.0-22.5) vs. 60.0 (31.0-87.5); p < 0.001). The parent's perception of the child's pain was significantly lower in the study group vs. the control group by the VAS (median (percentiles 25th-75th), 10.0 (10.0-20.0) vs. 50.0 (15.0-80.0); p < 0.001) and Wong-Baker Face scale (median (percentiles 25th-75th), 2.0 (2.0-2.0) vs. 4.0 (2.5-4.5); p < 0.001). The researcher's evaluation of the child's discomfort was also significantly lower in the study group (2.0 (1.0-3.0) vs. 4.0 (2.5-4.5); p < 0.001). The application of sonophoresis using SonoPrep® followed by the 5-min application of EMLA® cream showed significant benefit in young children in terms of pain reduction and patient satisfaction.