DATASET AI
<p dir="ltr">This dataset contains clinical, biological, and electrocardiographic parameters collected from adult patients diagnosed with ST-elevation myocardial infarction (STEMI), with the goal of supporting early prediction of cardiogenic shock (CS) during initial phases of care....
محفوظ في:
| المؤلف الرئيسي: | |
|---|---|
| منشور في: |
2025
|
| الموضوعات: | |
| الوسوم: |
إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
|
| الملخص: | <p dir="ltr">This dataset contains clinical, biological, and electrocardiographic parameters collected from adult patients diagnosed with ST-elevation myocardial infarction (STEMI), with the goal of supporting early prediction of cardiogenic shock (CS) during initial phases of care. The data were retrospectively collected from a single tertiary care center and include patients evaluated between the prehospital stage and cardiology-on-call consultation.</p><p dir="ltr">The dataset is organized into three distinct phases of early care:</p><ol><li><b>Prehospital phase</b> – includes variables recorded by emergency medical services (EMS) prior to hospital arrival.</li><li><b>Emergency department (ED) phase</b> – contains vital signs, initial labs, and ECG features available at ED presentation.</li><li><b>Cardiology-on-call phase</b> – reflects clinical reassessment prior to urgent coronary angiography.</li></ol><p dir="ltr">Key variables include Killip class, heart rate, systolic blood pressure, ECG rhythm abnormalities, creatinine, potassium, and other laboratory indicators of renal or hemodynamic dysfunction. These parameters were selected for their routine availability and clinical relevance in STEMI management.</p><p dir="ltr">The primary aim of this dataset is to enable the development and validation of machine learning models for:</p><ul><li>Early identification of STEMI patients at high risk of developing cardiogenic shock;</li><li>Clinical triage optimization and prioritization for urgent angiography;</li><li>Supporting time-sensitive decision-making in resource-limited or overcrowded emergency settings.</li></ul><p dir="ltr">The dataset includes model-ready variables suitable for classification tasks and has been used to train and evaluate algorithms such as Extra Trees, Support Vector Machines (SVM), Random Forest, and Quadratic Discriminant Analysis (QDA). Performance metrics include accuracy, precision, recall, F1-score, and Matthews Correlation Coefficient (MCC).</p><p dir="ltr">All data have been de-identified and processed in accordance with institutional ethical standards.</p> |
|---|