Hyperkalemia Detection in Emergency Departments Using Initial ECGs: A Smartphone AI ECG Analyzer vs. Board-Certified Physicians

Kim Donghoon,Jeong Joo,Kim Joonghee,Cho Youngjin,Park Inwon,Lee Sang-Min,Oh Young Taeck,Baek Sumin,Kang Dongin,Lee Eunkyoung,Jeong Bumi 대한의학회 2023 Journal of Korean medical science Vol.38 No.45

Background: Hyperkalemia is a potentially fatal condition that mandates rapid identification in emergency departments (EDs). Although a 12-lead electrocardiogram (ECG) can indicate hyperkalemia, subtle changes in the ECG often pose detection challenges. An artificial intelligence application that accurately assesses hyperkalemia risk from ECGs could revolutionize patient screening and treatment. We aimed to evaluate the efficacy and reliability of a smartphone application, which utilizes camera-captured ECG images, in quantifying hyperkalemia risk compared to human experts. Methods: We performed a retrospective analysis of ED hyperkalemic patients (serum potassium ≥ 6 mmol/L) and their age- and sex-matched non-hyperkalemic controls. The application was tested by five users and its performance was compared to five board-certified emergency physicians (EPs). Results: Our study included 125 patients. The area under the curve (AUC)-receiver operating characteristic of the application’s output was nearly identical among the users, ranging from 0.898 to 0.904 (median: 0.902), indicating almost perfect interrater agreement (Fleiss’ kappa 0.948). The application demonstrated high sensitivity (0.797), specificity (0.934), negative predictive value (NPV) (0.815), and positive predictive value (PPV) (0.927). In contrast, the EPs showed moderate interrater agreement (Fleiss’ kappa 0.551), and their consensus score had a significantly lower AUC of 0.662. The physicians’ consensus demonstrated a sensitivity of 0.203, specificity of 0.934, NPV of 0.527, and PPV of 0.765. Notably, this performance difference remained significant regardless of patients’ sex and age (P < 0.001 for both). Conclusion: Our findings suggest that a smartphone application can accurately and reliably quantify hyperkalemia risk using initial ECGs in the ED.

A Retrospective Clinical Evaluation of an Artificial Intelligence Screening Method for Early Detection of STEMI in the Emergency Department

Kim Dongsung,Hwang Ji Eun,Cho Youngjin,Cho Hyoung-Won,Lee Wonjae,Lee Ji Hyun,Oh Il-Young,Baek Sumin,Lee Eunkyoung,Kim Joonghee 대한의학회 2022 Journal of Korean medical science Vol.37 No.10

Background: Rapid revascularization is the key to better patient outcomes in ST-elevation myocardial infarction (STEMI). Direct activation of cardiac catheterization laboratory (CCL) using artificial intelligence (AI) interpretation of initial electrocardiography (ECG) might help reduce door-to-balloon (D2B) time. To prove that this approach is feasible and beneficial, we assessed the non-inferiority of such a process over conventional evaluation and estimated its clinical benefits, including a reduction in D2B time, medical cost, and 1-year mortality. Methods: This is a single-center retrospective study of emergency department (ED) patients suspected of having STEMI from January 2021 to June 2021. Quantitative ECG (QCG™), a comprehensive cardiovascular evaluation system, was used for screening. The non-inferiority of the AI-driven CCL activation over joint clinical evaluation by emergency physicians and cardiologists was tested using a 5% non-inferiority margin. Results: Eighty patients (STEMI, 54 patients [67.5%]) were analyzed. The area under the curve of QCG score was 0.947. Binned at 50 (binary QCG), the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were 98.1% (95% confidence interval [CI], 94.6%, 100.0%), 76.9% (95% CI, 60.7%, 93.1%), 89.8% (95% CI, 82.1%, 97.5%) and 95.2% (95% CI, 86.1%, 100.0%), respectively. The difference in sensitivity and specificity between binary QCG and the joint clinical decision was 3.7% (95% CI, −3.5%, 10.9%) and 19.2% (95% CI, −4.7%, 43.1%), respectively, confirming the non-inferiority. The estimated median reduction in D2B time, evaluation cost, and the relative risk of 1-year mortality were 11.0 minutes (interquartile range [IQR], 7.3–20.0 minutes), 26,902.2 KRW (22.78 USD) per STEMI patient, and 12.39% (IQR, 7.51–22.54%), respectively. Conclusion: AI-assisted CCL activation using initial ECG is feasible. If such a policy is implemented, it would be reasonable to expect some reduction in D2B time, medical cost, and 1-year mortality.

지하철 기관사 휴게 공간 디자인을 위한 요구조사 분석

김국선(Kooksun Kim),김민지(Minji Kim),김중희(Joonghee Kim),송정효(Jeonghyo Song),신중배(Jungbae Shin),윤종현(Jonghyun Yun),최보아(Boah Choi) 한국가구학회 2015 한국가구학회지 Vol.26 No.1

There is an increasing trend each year in the number of subway users due to increase in the traffic volume. As the number of train service has risen, the subway drivers are given heavy work and from the same, are put at risk of mental health deterioration due to stress. As this is directly connected to safety, there is an urgent need for improvement of the situation. Therefore, as a way to appropriately relieve such stress, resting space where mental and physical energy can be recharged is necessary. Thus, this research conducted a survey of demand with subway drivers as subjects in order to provide resting space that can offer healthy rest and increase work effectiveness. 50 subway drivers were the subjects of this survey. As a result of analyzing the contents of the survey, environmental satisfaction was shown to be the lowest compared to psychological and morphological satisfaction. In the case of sleeping room, they wanted more comfortable and individual space. And the waiting room had the highest rate of usage frequency. The storage space was where the subway drivers showed the strongest attachment and was shown to be the only personal space. Most were shown to be satisfied with the outdoor space, but it had the lowest usage frequency. In addition, it is nessasery customized storage cabinet designed for subway drivers should be produced, which can effectively arrange the limited space. I hope that these direction of improvement can increase the healthy rest and work effectiveness of the locomotive engineers as it is directly being connected to the health and safety of the citizens.

구조기구조기인 아이들 소음 저감을 위한 차체 골격 내 복합체 삽입 최적설계

김효식(Hyosig Kim),윤성호(Seongho Yoon),김중희(Joonghee Kim) 한국자동차공학회 2010 한국자동차공학회 부문종합 학술대회 Vol.2010 No.5

As a matter of fact, it has been difficult to modify the shape of vehicle body skeletons since the technical definition for a global body structure was fixed and the corresponding molds were developed. By the way, if it is available to apply an alternative to reinforce the skeletons without changing their molds, it will be more flexible to improve the performance qualities relevant to crash, durability and NVH(Noise, Vibration and Harshness). Recently, a solution with the composite body becomes available for the purpose. We present an optimal design method to insert the composite bodies into a vehicle body skeleton in order to improve the structure-borne noise at the idle condition. The algorithms, topology optimization and gradient based optimization, are applied to mainly search the sensitive structural joints in the body skeleton and to extract the technical definition of composite body insertions. Inserting the composite bodies into the sensitive joints, it is predicted to achieve the countermeasure which can meet the design availability in terms of the idle noise and the weight. According to the validation result with test vehicles, the concerned noise transfer function is reduced and the idle booming noise is resultantly improved.

Blood pressure–targeted stepwise resuscitation for hemorrhagic shock in rats

Lee, Jae Hyuk,Kim, Kyuseok,Jo, You Hwan,Kim, Min A,Lee, Kyoung-Bun,Rhee, Joong Eui,Doo, Ah-Reum,Lee, Min Ji,Park, Chan Jong,Kim, Joonghee,Chung, Heajin Lippincott Williams Wilkins, Inc. 2014 The journal of trauma and acute care surgery Vol.76 No.3

BACKGROUND: Generation of reactive oxygen species (ROS) is an important mechanism of ischemia-reperfusion injury. Abrupt reoxygenation compared with slow reoxygenation has been known to increase ROS generation. Thus, slow and stepwise reperfusion can reduce ROS generation and subsequent ischemia-reperfusion injury. This study investigated the effect of slow reperfusion by blood pressure–targeted stepwise resuscitation (PSR) in hemorrhagic shock. METHODS: Pressure-controlled hemorrhagic shock was induced in male Sprague-Dawley rats for 1 hour. Rats were then allocated to one of three groups (no-resuscitation group, n = 14; PSR group, n = 15; rapid normalization of blood pressure (RR) group, n = 15). Survival time and hemodynamic changes were recorded and compared. Blood samples and liver tissue were harvested after 6 hours of resuscitation in surviving rats. RESULTS: All of the rats in the no-resuscitation group were expired before the end of the 6-hour observation period. Survival times were significantly longer in the PSR group than in the RR group (survival rates, 11 of 15 vs. 5 of 15, log rank p = 0.032). Plasma amino alanine transferase, histologic liver injury, and ROS generation in the liver tissue were significantly lower in the PSR group than in the RR group (all findings significant, p < 0.05). In addition, PSR significantly decreased plasma nitric oxide, liver interleukin 1&bgr;, and liver interleukin 6 compared with rapid resuscitation in addition to augmenting Akt survival pathways (all p < 0.05). CONCLUSION: Slow reperfusion by PSR decreased mortality, ROS generation, and liver injury in rats undergoing hemorrhagic shock. Stepwise resuscitation also decreased inflammatory cytokine production and augmented Akt survival pathways.

Prediction of Neurological Outcomes in Out-of-hospital Cardiac Arrest Survivors Immediately after Return of Spontaneous Circulation: Ensemble Technique with Four Machine Learning Models

Heo Ji Han,Kim Taegyun,Shin Jonghwan,Suh Gil Joon,Kim Joonghee,Jung Yoon Sun,Park Seung Min,Kim Sungwan 대한의학회 2021 Journal of Korean medical science Vol.36 No.28

Background: We performed this study to establish a prediction model for 1-year neurological outcomes in out-of-hospital cardiac arrest (OHCA) patients who achieved return of spontaneous circulation (ROSC) immediately after ROSC using machine learning methods. Methods: We performed a retrospective analysis of an OHCA survivor registry. Patients aged ≥ 18 years were included. Study participants who had registered between March 31, 2013 and December 31, 2018 were divided into a develop dataset (80% of total) and an internal validation dataset (20% of total), and those who had registered between January 1, 2019 and December 31, 2019 were assigned to an external validation dataset. Four machine learning methods, including random forest, support vector machine, ElasticNet and extreme gradient boost, were implemented to establish prediction models with the develop dataset, and the ensemble technique was used to build the final prediction model. The prediction performance of the model in the internal validation and the external validation dataset was described with accuracy, area under the receiver-operating characteristic curve, area under the precision-recall curve, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Futhermore, we established multivariable logistic regression models with the develop set and compared prediction performance with the ensemble models. The primary outcome was an unfavorable 1-year neurological outcome. Results: A total of 1,207 patients were included in the study. Among them, 631, 139, and 153 were assigned to the develop, the internal validation and the external validation datasets, respectively. Prediction performance metrics for the ensemble prediction model in the internal validation dataset were as follows: accuracy, 0.9620 (95% confidence interval [CI], 0.9352–0.9889); area under receiver-operator characteristics curve, 0.9800 (95% CI, 0.9612– 0.9988); area under precision-recall curve, 0.9950 (95% CI, 0.9860–1.0000); sensitivity, 0.9594 (95% CI, 0.9245–0.9943); specificity, 0.9714 (95% CI, 0.9162–1.0000); PPV, 0.9916 (95% CI, 0.9752–1.0000); NPV, 0.8718 (95% CI, 0.7669–0.9767). Prediction performance metrics for the model in the external validation dataset were as follows: accuracy, 0.8509 (95% CI, 0.7825–0.9192); area under receiver-operator characteristics curve, 0.9301 (95% CI, 0.8845–0.9756); area under precision-recall curve, 0.9476 (95% CI, 0.9087–0.9867); sensitivity, 0.9595 (95% CI, 0.9145–1.0000); specificity, 0.6500 (95% CI, 0.5022–0.7978); PPV, 0.8353 (95% CI, 0.7564–0.9142); NPV, 0.8966 (95% CI, 0.7857–1.0000). All the prediction metrics were higher in the ensemble models, except NPVs in both the internal and the external validation datasets. Conclusion: We established an ensemble prediction model for prediction of unfavorable 1-year neurological outcomes in OHCA survivors using four machine learning methods. The prediction performance of the ensemble model was higher than the multivariable logistic regression model, while its performance was slightly decreased in the external validation dataset.

공회전시 배기계 이음에 대한 수학적 계산 및 예측기법에 대한 연구

장진만(Jinman Jang),김기동(Kidong Kim),김준완(Junewan Kim),김증희(Joonghee Kim) 한국자동차공학회 2011 한국자동차공학회 학술대회 및 전시회 Vol.2011 No.11

In general, the main source of vehicle noise is the ignition of engine and to diminish the degree of the noise, exhaust system is used as complicated filters for attenuating the noise more or less. But, even these kinds of filters are playing their roles at their expected frequencies bands, some noises still can be detected by passengers. They are so called ‘abnormal noises’ ,which are discovered when their anti-filters are working well, and it is very difficult to find what is the real problem and what is the basis of this phenomenon. Usually, frequency analysis is the most useful method for studying noise phenomena, but main effect of abnormal noise comes from listener’s feeling, therefore only frequency analysis process is not enough to clarify not only the cause but also the phenomenon itself. In this study, one of the abnormal noise, which can be detected very easily at the tail pipe, has been examined by mathematical way to define the cause of feeling of abnormal noise more effectively and precisely, furthermore also noise filter development process has been developed as a result in order to design proper filter system for exhaust system. For the validation of this calculation and design process, virtual noise has been exerted to calculated filters which had been designed properly by the calculation of former step and then, the noise has been accessed as a listener’s feeling by sound feeling matrices of octave band modulation.

점성-대류 음향 방정식과 비 도플러 알고리즘을 이용한 배기계의 소음 성능 예측에 관한 연구

장진만(Jinman Jang),강영춘(Youngchun Kang),김기동(Kidong Kim),김준완(Junewan Kim),김영수(Youngsoo Kim),김중희(Joonghee Kim) 한국자동차공학회 2010 한국자동차공학회 학술대회 및 전시회 Vol.2010 No.11

Recently, the noise of vehicle is the one of the key factors for customers to purchase a vehicle and the most important part which is related to the noise is the exhaust system. Thus, car makers have their own ways to assess this exhaust noise not only to decrease the level of noise but to enhance the feeling of it. Typically, to do these things in the early stage of development, the tuning code of the exhaust system has to be made by CAE tool, which is very reliable but expensive, and the prototype parts of this code would be made for the validation test. Then this process can be iterated to meet the target of the performance. In this study, a new algorithm which adapts the ‘convective sound wave theory’ and ‘anti-Doppler effect’ has been introduced and with this algorithm, a calculation code for sound field estimation has been developed. And for the validation of the calculation code, various comparisons and validations were carried out with the result of wind tunnel test. Also the comparison with other CAE tool has been performed for the validity and the improvement of the new calculation code could be achieved.

엔진 마운팅 및 배기 시스템의 동역학적 특성계발을 위한 시스템 벤치의 연구

장진만(Jinman Jang),강영춘(Youngchun Kang),김준완(Junewan Kim),김영수(Youngsoo Kim),김중희(Joonghee Kim) 한국자동차공학회 2010 한국자동차공학회 부문종합 학술대회 Vol.2010 No.5

It is very important to understand the dynamic characteristics of a vehicle like idle vibration, key on-off, and engine shake in the early stage of the projects before the prototype could not be prepared yet. Also, if there is no enough validation on the design of these performances, this can cause large changes in the engineering later on. Therefore, it is worthwhile to study a new method to validate the dynamic performances in early stage of engineering before advancing to the next stages. In this study, a system bench for the low frequency vibration simulation is introduced for the prediction and validation of power train and exhaust system dynamic performances. The system bench was built to have equivalent dynamic characteristics of the vehicle and power train system, and various comparisons and validations were carried out with CAE and vehicle test results so that validity and improvement points of the bench could be achieved.

DERIVATION AND VALIDATION OF NONLINEAR GOVERNING EQUATIONS FOR THE ANALYSIS OF VEHICLE HANDLING

SungMo YANG,JoongHee KIM 한국자동차공학회 2010 한국자동차공학회 학술대회 및 전시회 Vol.2010 No.11

Nonlinear governing equations for the handling analysis of a ground vehicle are derived by the Lagrange equations of motion. The derived equations are coded using VBA (Visual Basic for Applications) embedded in Microsoft’s Excel Software and simulated in the time domain by the 4th-order Runge-Kutta method. A total six degrees of freedom in the equations are used: Three of these are the directional translation, lateral translation and yaw of a platform (unsprung) on the base of an inertial ground coordinate and the other three are the roll, pitch, and yaw of a body (sprung) by a platform-fixed coordinate. Four driving torques and four wheel angles of all tires are used as input control parameters. A simplified Calspan tire model is adopted for generalized forces of equations. This is a combined model which can obtain tractional (or braking) and side forces by the inputs of the directional and side slip ratios and the vertical force. The VBA code realized in this study is validated by comparisons with the static equilibrium results and the test data cited in published papers. The major characteristics of this paper are: (1) the coordinate systems of equations are mixed with the inertial frame and the platformfixed frame, and as a result almost all kinds of driving conditions with large mileages are possible to simulate (2) vertical movement is eliminated due to focusing on the handling analysis (3) the body-yaw dof is separated from the platform-yaw dof and (4) lastly, the programming is done by VBA, which is rarely used in the vehicle dynamics field.