Data Mining Technology Combined with Out-of-Hospital Cardiac Arrest, Symptom Association and Prediction Model Probing

2018 ◽  
pp. 298-309
Author(s):  
Chih-Chun Chang ◽  
Jui-Hung Kao ◽  
Chien-Yeh Hsu ◽  
Homg-Twu Liaw ◽  
Tse-Chun Wang
Keyword(s):  
Data Mining ◽  
Cardiac Arrest ◽  
Prediction Model ◽  
Mining Technology ◽  
Hospital Cardiac Arrest

Abstract 354: A Prediction Model for Patients With Emergency Medical Technicians Witnessed Out-of-Hospital Cardiac Arrest

Circulation ◽  
2018 ◽  
Vol 138 (Suppl_2) ◽  
Author(s):  
Ming-Ju Hsieh ◽  
Wen-Chu Chiang ◽  
Wei-Tien Chang ◽  
Chih-Wei Yang ◽  
Yu-Chun Chien ◽  
...  
Keyword(s):  
Cardiac Arrest ◽  
Prediction Model ◽  
Situation Awareness ◽  
Blood Pressure Level ◽  
Emergency Medical Technicians ◽  
Validation Dataset ◽  
Systolic Blood Pressure Level ◽  
Value System ◽  
Emergency Medical ◽  
Hospital Cardiac Arrest

Introduction: In-hospital early warning system scores for prediction of clinical deterioration have been well-developed. However, such prediction tools in prehospital setting remain unavailable. Hypothesis: To develop a model for predicting patients with emergency medical technicians witnessed out-of-hospital cardiac arrest (EMT-witnessed OHCA) . Methods: We used the fire-based emergency medical service (EMS) data from Taipei city to develop the prediction model. Patients included in this study were those initially alive, non-traumatic, and aged ≧20 years. Data were extracted from records of ambulance run sheets and OHCA registry in Taipei. The primary outcome (i.e. EMT-witnessed OHCA) was defined as cardiac arrest occurring during EMT services before arrival at the receiving hospital. The prediction model was developed through the standard cross-validation method (i.e. divided dataset for training group and validation group). Area under the receiver operating characteristic curve (AUC) and Hosmer-Lemeshow (HL) test were used to test discrimination and calibration. The point value system with Youden’s J Index was used to find the best cut-off value for practical application. Results: From 2011 to 2015, a total of 252,771 patients were included. Of them, 660 (0.26%) were EMT-witnessed OHCA. The prediction model, including gender, respiratory rate, heart rate, systolic blood pressure, level of consciousness and oxygen saturation, showed excellent discrimination (AUC 0.94) and calibration ( p =0.42 for HL test). When applied to the validation dataset, it maintained good discriminatory ability (AUC 0.94) and calibration ( p =0.11). The optimal cut-off value (≧13) of the point value system of the tool showed high sensitivity (87.84%) and specificity (86.20%). Conclusions: The newly developed prediction model will help identify high-risk patients with EMT-witnessed OHCA and indicate potential prevention by situation awareness in EMS.

Spatial analysis and data mining techniques for identifying risk factors of Out-of-Hospital Cardiac Arrest

2017 ◽  
Vol 37 (1) ◽  
pp. 1528-1538 ◽  
Author(s):  
Jui-Hung Kao ◽  
Ta-Chien Chan ◽  
Feipei Lai ◽  
Bo-Cheng Lin ◽  
Wei-Zen Sun ◽  
...  
Keyword(s):  
Risk Factors ◽  
Data Mining ◽  
Cardiac Arrest ◽  
Spatial Analysis ◽  
Data Mining Techniques ◽  
Hospital Cardiac Arrest

scholarly journals Developing a time-adaptive prediction model for out-of-hospital cardiac arrest: results from a nationwide cohort study in Korea (Preprint)

2021 ◽  
Author(s):  
Ji Woong Kim ◽  
Juhyung Ha ◽  
Taerim Kim ◽  
Hee Yoon ◽  
Sung Yeon Hwang ◽  
...  
Keyword(s):  
Cardiac Arrest ◽  
Prediction Model ◽  
Characteristic Curve ◽  
Model Performance ◽  
Spontaneous Circulation ◽  
Training Dataset ◽  
Public Health Issue ◽  
Adaptive Prediction ◽  
Prediction Probability ◽  
Hospital Cardiac Arrest

BACKGROUND Out-of-hospital cardiac arrest (OHCA) is a serious public health issue and predicting the prognosis of OHCA patients would be helpful for clinicians to make decisions on the treatment of patients or use of hospital resources. OBJECTIVE This study aimed to develop a time-adaptive conditional prediction model (TACOM) for predicting clinical outcomes every minute. METHODS We performed a retrospective observational study using data from the Korea OHCA Registry in South Korea. In this study, we excluded patients with trauma, those who experienced return of spontaneous circulation before arriving in the emergency departments (ED), and those who did not receive cardiopulmonary resuscitation (CPR) in the ED. We selected patients who received CPR in the ED. To develop the time-adaptive prediction model, we organized the training dataset as ongoing CPR patients by the minute. We used LightGBM as the machine-learning method. Model performance was quantified using the prediction probability of the model, area under the receiver operating characteristic curve (AUROC), and area under the precision recall curve. RESULTS From 0 to 30 min, the AUROC of TACOM for predicting good neurologic outcomes ranged from 0.910 (0.910–0.911) to 0.869 [0.865–0.871], whereas that for survival to hospital discharge was 0.800 (0.797–0.800) to 0.734 (0.736–0.740). The prediction probability of TACOM showed similar flow with cohort data, based on a comparison with the conventional model’s prediction probability. CONCLUSIONS TACOM predicted the clinical outcome of patients with OHCA every minute. This model for predicting patient outcomes by minute can be expected to help clinicians make rational decisions for OHCA patients.

scholarly journals Online English Teaching Course Score Analysis Based on Decision Tree Mining Algorithm

Complexity ◽  
2021 ◽  
Vol 2021 ◽  
pp. 1-10
Author(s):  
Xiaojun Jiang
Keyword(s):  
Data Mining ◽  
Prediction Model ◽  
Warning System ◽  
College Teaching ◽  
Information Value ◽  
Retention Prediction ◽  
Minimum Problem ◽  
Mining Technology ◽  
Teaching Management ◽  
Advance Student

With the advent of the Big Data era, information and data are growing in spurts, fueling the deep application of information technology in all levels of society. It is especially important to use data mining technology to study the industry trends behind the data and to explore the information value contained in the massive data. As teaching and learning in higher education continue to advance, student academic and administrative data are growing at a rapid pace. In this paper, we make full use of student academic data and campus behavior data to analyze the data inherent patterns and correlations and use these patterns rationally to provide guidance for teaching activities and teaching management, thus further improving the quality of teaching management. The establishment of a data-mining-technology-based college repetition warning system can help student management departments to strengthen supervision, provide timely warning information for college teaching management as well as leaders and counselors’ decision-making, and thus provide early help to students with repetition warnings. In this paper, we use the global search advantage of genetic algorithm to build a GABP hybrid prediction model to solve the local minimum problem of BP neural network algorithm. The data validation results show that Recall reaches 95% and F1 result is about 86%, and the accuracy of the algorithm prediction results is improved significantly. It can provide a solid data support basis for college administrators to predict retention. Finally, the problems in the application of the retention prediction model are analyzed and corresponding suggestions are given.

scholarly journals Spatio-temporal prediction model of out-of-hospital cardiac arrest: Designation of medical priorities and estimation of human resources requirement

PLoS ONE ◽  
2020 ◽  
Vol 15 (8) ◽  
pp. e0238067
Author(s):  
Angelo Auricchio ◽  
Stefano Peluso ◽  
Maria Luce Caputo ◽  
Jost Reinhold ◽  
Claudio Benvenuti ◽  
...  
Keyword(s):  
Cardiac Arrest ◽  
Human Resources ◽  
Prediction Model ◽  
Temporal Prediction ◽  
Spatio Temporal ◽  
Hospital Cardiac Arrest
Sign in / Sign up

Export Citation Format

Share Document