Importance: For patients with chronic kidney disease (CKD), hyperkalemia is common, associated with fatal arrhythmias, and often asymptomatic, while guideline-directed monitoring of serum potassium is underused. A deep-learning model that enables noninvasive hyperkalemia screening from the electrocardiogram (ECG) may improve detection of this life-threatening condition. Objective: To evaluate the performance of a deep-learning model in detection of hyperkalemia from the ECG in patients with CKD. Design, Setting, and Participants: A deep convolutional neural network (DNN) was trained using 1 576 581 ECGs from 449 380 patients seen at Mayo Clinic, Rochester, Minnesota, from 1994 to 2017. The DNN was trained using 2 (leads I and II) or 4 (leads I, II, V3, and V5) ECG leads to detect serum potassium levels of 5.5 mEq/L or less (to convert to millimoles per liter, multiply by 1) and was validated using retrospective data from the Mayo Clinic in Minnesota, Florida, and Arizona. The validation included 61 965 patients with stage 3 or greater CKD. Each patient had a serum potassium count drawn within 4 hours after their ECG was recorded. Data were analyzed between April 12, 2018, and June 25, 2018. Exposures: Use of a deep-learning model. Main Outcomes and Measures: Area under the receiver operating characteristic curve (AUC) and sensitivity and specificity, with serum potassium level as the reference standard. The model was evaluated at 2 operating points, 1 for equal specificity and sensitivity and another for high (90%) sensitivity. Results: Of the total 1 638 546 ECGs, 908 000 (55%) were from men. The prevalence of hyperkalemia in the 3 validation data sets ranged from 2.6% (n = 1282 of 50 099; Minnesota) to 4.8% (n = 287 of 6011; Florida). Using ECG leads I and II, the AUC of the deep-learning model was 0.883 (95% CI, 0.873-0.893) for Minnesota, 0.860 (95% CI, 0.837-0.883) for Florida, and 0.853 (95% CI, 0.830-0.877) for Arizona. Using a 90% sensitivity operating point, the sensitivity was 90.2% (95% CI, 88.4%-91.7%) and specificity was 63.2% (95% CI, 62.7%-63.6%) for Minnesota; the sensitivity was 91.3% (95% CI, 87.4%-94.3%) and specificity was 54.7% (95% CI, 53.4%-56.0%) for Florida; and the sensitivity was 88.9% (95% CI, 84.5%-92.4%) and specificity was 55.0% (95% CI, 53.7%-56.3%) for Arizona. Conclusions and Relevance: In this study, using only 2 ECG leads, a deep-learning model detected hyperkalemia in patients with renal disease with an AUC of 0.853 to 0.883. The application of artificial intelligence to the ECG may enable screening for hyperkalemia. Prospective studies are warranted.
Importance: For patients with chronic kidney disease (CKD), hyperkalemia is common, associated with fatal arrhythmias, and often asymptomatic, while guideline-directed monitoring of serum potassium is underused. A deep-learning model that enables noninvasive hyperkalemia screening from the electrocardiogram (ECG) may improve detection of this life-threatening condition. Objective: To evaluate the performance of a deep-learning model in detection of hyperkalemia from the ECG in patients with CKD. Design, Setting, and Participants: A deep convolutional neural network (DNN) was trained using 1 576 581 ECGs from 449 380 patients seen at Mayo Clinic, Rochester, Minnesota, from 1994 to 2017. The DNN was trained using 2 (leads I and II) or 4 (leads I, II, V3, and V5) ECG leads to detect serum potassium levels of 5.5 mEq/L or less (to convert to millimoles per liter, multiply by 1) and was validated using retrospective data from the Mayo Clinic in Minnesota, Florida, and Arizona. The validation included 61 965 patients with stage 3 or greater CKD. Each patient had a serum potassium count drawn within 4 hours after their ECG was recorded. Data were analyzed between April 12, 2018, and June 25, 2018. Exposures: Use of a deep-learning model. Main Outcomes and Measures: Area under the receiver operating characteristic curve (AUC) and sensitivity and specificity, with serum potassium level as the reference standard. The model was evaluated at 2 operating points, 1 for equal specificity and sensitivity and another for high (90%) sensitivity. Results: Of the total 1 638 546 ECGs, 908 000 (55%) were from men. The prevalence of hyperkalemia in the 3 validation data sets ranged from 2.6% (n = 1282 of 50 099; Minnesota) to 4.8% (n = 287 of 6011; Florida). Using ECG leads I and II, the AUC of the deep-learning model was 0.883 (95% CI, 0.873-0.893) for Minnesota, 0.860 (95% CI, 0.837-0.883) for Florida, and 0.853 (95% CI, 0.830-0.877) for Arizona. Using a 90% sensitivity operating point, the sensitivity was 90.2% (95% CI, 88.4%-91.7%) and specificity was 63.2% (95% CI, 62.7%-63.6%) for Minnesota; the sensitivity was 91.3% (95% CI, 87.4%-94.3%) and specificity was 54.7% (95% CI, 53.4%-56.0%) for Florida; and the sensitivity was 88.9% (95% CI, 84.5%-92.4%) and specificity was 55.0% (95% CI, 53.7%-56.3%) for Arizona. Conclusions and Relevance: In this study, using only 2 ECG leads, a deep-learning model detected hyperkalemia in patients with renal disease with an AUC of 0.853 to 0.883. The application of artificial intelligence to the ECG may enable screening for hyperkalemia. Prospective studies are warranted.
Authors: Luis Antonio Salazar-Licea; Jesús Carlos Pedraza-Ortega; Alberto Pastrana-Palma; Marco A Aceves-Fernandez Journal: Comput Methods Programs Biomed Date: 2017-02-24 Impact factor: 5.428
Authors: Omar Z Yasin; Zachi Attia; John J Dillon; Christopher V DeSimone; Yehu Sapir; Jennifer Dugan; Virend K Somers; Michael J Ackerman; Samuel J Asirvatham; Christopher G Scott; Kevin E Bennet; Dorothy J Ladewig; Dan Sadot; Amir B Geva; Paul A Friedman Journal: J Electrocardiol Date: 2017-06-08 Impact factor: 1.438
Authors: Venu Velagapudi; John C O'Horo; Anu Vellanki; Stephen P Baker; Rahul Pidikiti; Jeffrey S Stoff; Dennis A Tighe Journal: J Electrocardiol Date: 2016-09-09 Impact factor: 1.438
Authors: David N Juurlink; Muhammad M Mamdani; Douglas S Lee; Alexander Kopp; Peter C Austin; Andreas Laupacis; Donald A Redelmeier Journal: N Engl J Med Date: 2004-08-05 Impact factor: 91.245
Authors: Patrick Rossignol; Daniela Dobre; John J V McMurray; Karl Swedberg; Henry Krum; Dirk J van Veldhuisen; Harry Shi; Michael Messig; John Vincent; Nicolas Girerd; George Bakris; Bertram Pitt; Faiez Zannad Journal: Circ Heart Fail Date: 2013-12-02 Impact factor: 8.790
Authors: Zachi I Attia; Christopher V DeSimone; John J Dillon; Yehu Sapir; Virend K Somers; Jennifer L Dugan; Charles J Bruce; Michael J Ackerman; Samuel J Asirvatham; Bryan L Striemer; Jan Bukartyk; Christopher G Scott; Kevin E Bennet; Dorothy J Ladewig; Emily J Gilles; Dan Sadot; Amir B Geva; Paul A Friedman Journal: J Am Heart Assoc Date: 2016-01-25 Impact factor: 5.501
Authors: Albert K Feeny; Mina K Chung; Anant Madabhushi; Zachi I Attia; Maja Cikes; Marjan Firouznia; Paul A Friedman; Matthew M Kalscheur; Suraj Kapa; Sanjiv M Narayan; Peter A Noseworthy; Rod S Passman; Marco V Perez; Nicholas S Peters; Jonathan P Piccini; Khaldoun G Tarakji; Suma A Thomas; Natalia A Trayanova; Mintu P Turakhia; Paul J Wang Journal: Circ Arrhythm Electrophysiol Date: 2020-07-06
Authors: Albert K Feeny; John Rickard; Kevin M Trulock; Divyang Patel; Saleem Toro; Laurie Ann Moennich; Niraj Varma; Mark J Niebauer; Eiran Z Gorodeski; Richard A Grimm; John Barnard; Anant Madabhushi; Mina K Chung Journal: Circ Arrhythm Electrophysiol Date: 2020-06-14