The purpose of this study is to validate the real-world performance of a previously developed Artificial Intelligence - Electrocardiogram (AI-ECG) algorithm for identification of hyperkalemia with a six-lead mobile-enhanced device .
1. Ambulatory adult patients in the Emergency Department (ED) at increased risk for hyperkalemia (due to age ≥ 50 years, and one or more criteria including estimated Glomerular filtration rate (eGFR) (from serum creatinine) \< 45 ml/minute and/or a history of serum potassium \> 5.2 milliequivalents per liter (mEq/l) who present to the emergency department will be approached to consent for the rapid screening process. 2. Those who consent will undergo 30 second 6 L ECG recording with a portable, mobile-enhanced device (AliveCor Kardia). 3. This ECG data is subsequently evaluated by our artificial intelligence algorithm to detect hyperkalemia, and the estimated probability of hyperkalemia is recorded. 4. The research team notifies supervising Emergency Department staff of patients whose probability of hyperkalemia is significantly elevated above the optimized cutoff point according to the AI-ECG algorithm.
Study Type
OBSERVATIONAL
Enrollment
1,151
Mayo Clinic
Rochester, Minnesota, United States
Hyperkalemia detection by AI enhanced ECG
Understanding model's ability to predict hyperkalemia as determined by the area under the receiver operating characteristic
Time frame: 12 months
Performance metrics for the detection of hyperkalemia by AI enhanced ECG
Detailed performance metrics of the algorithm (sensitivity, specificity, positive predictive value and negative predictive value) will be calculated using an optimized cutoff threshold determined from the primary outcome.
Time frame: 12 months
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.