Noninvasive assessment of dofetilide plasma concentration using a deep learning (neural network) analysis of the surface electrocardiogram: A proof of concept study

Background Dofetilide is an effective antiarrhythmic medication for rhythm control in atrial fibrillation, but carries a significant risk of pro-arrhythmia and requires meticulous dosing and monitoring. The cornerstone of this monitoring, measurement of the QT/QTc interval, is an imperfect surrogate for plasma concentration, efficacy, and risk of pro-arrhythmic potential. Objective The aim of our study was to test the application of a deep learning approach (using a convolutional neural network) to assess morphological changes on the surface ECG (beyond the QT interval) in relation to dofetilide plasma concentrations. Methods We obtained publically available serial ECGs and plasma drug concentrations from 42 healthy subjects who received dofetilide or placebo in a placebo‐controlled cross‐over randomized controlled clinical trial. Three replicate 10-s ECGs were extracted at predefined time-points with simultaneous measurement of dofetilide plasma concentration We developed a deep learning algorithm to predict dofetilide plasma concentration in 30 subjects and then tested the model in the remaining 12 subjects. We compared the deep leaning approach to a linear model based only on QTc. Results Fourty two healthy subjects (21 females, 21 males) were studied with a mean age of 26.9 ± 5.5 years. A linear model of the QTc correlated reasonably well with dofetilide drug levels (r = 0.64). The best correlation to dofetilide level was achieved with the deep learning model (r = 0.85). Conclusion This proof of concept study suggests that artificial intelligence (deep learning/neural network) applied to the surface ECG is superior to analysis of the QT interval alone in predicting plasma dofetilide concentration.

[1]  I. Kohane,et al.  Big Data and Machine Learning in Health Care. , 2018, JAMA.

[2]  Michael Christiansen,et al.  T wave morphology analysis distinguishes between KvLQT1 and HERG mutations in long QT syndrome. , 2004, Heart rhythm.

[3]  C. Funck-Brentano,et al.  Influence of dofetilide on QT-interval duration and dispersion at various heart rates during exercise in humans. , 1996, Circulation.

[4]  Naif Alajlan,et al.  Deep learning approach for active classification of electrocardiogram signals , 2016, Inf. Sci..

[5]  P. Noseworthy,et al.  Electrocardiographic Predictors of Torsadogenic Risk During Dofetilide or Sotalol Initiation: Utility of a Novel T Wave Analysis Program , 2015, Cardiovascular Drugs and Therapy.

[6]  Yoshua Bengio,et al.  Understanding the difficulty of training deep feedforward neural networks , 2010, AISTATS.

[7]  A. Camm,et al.  Dofetilide in patients with congestive heart failure and left ventricular dysfunction. Danish Investigations of Arrhythmia and Mortality on Dofetilide Study Group. , 1999, The New England journal of medicine.

[8]  D. Walker,et al.  Pharmacokinetic and pharmacodynamic effects of UK-68,798, a new potential class III antiarrhythmic drug. , 1991, British journal of clinical pharmacology.

[9]  R. Woosley,et al.  Bazett and Fridericia QT correction formulas interfere with measurement of drug-induced changes in QT interval. , 2006, Heart rhythm.

[10]  Marek Malik,et al.  Errors and misconceptions in ECG measurement used for the detection of drug induced QT interval prolongation. , 2004, Journal of electrocardiology.

[11]  Lei Wang,et al.  Automated atrial fibrillation detection based on deep learning network , 2016, 2016 IEEE International Conference on Information and Automation (ICIA).

[12]  U. Rajendra Acharya,et al.  Application of deep convolutional neural network for automated detection of myocardial infarction using ECG signals , 2017, Inf. Sci..

[13]  P. Noseworthy,et al.  Identification of Concealed and Manifest Long QT Syndrome Using a Novel T Wave Analysis Program , 2016, Circulation. Arrhythmia and electrophysiology.

[14]  B. Nørgaard,et al.  Efficacy and safety of intravenously administered dofetilide in acute termination of atrial fibrillation and flutter: A multicenter, randomized, double-blind, placebo-controlled trial , 1999 .

[15]  Norman Stockbridge,et al.  Comprehensive T wave Morphology Assessment in a Randomized Clinical Study of Dofetilide, Quinidine, Ranolazine, and Verapamil , 2015, Journal of the American Heart Association.

[16]  Jian Sun,et al.  Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[17]  C. Funck-Brentano,et al.  Pharmacokinetic and pharmacodynamic modeling of the effects of oral and intravenous administrations of dofetilide on ventricular repolarization , 1995, Clinical pharmacology and therapeutics.

[18]  M. Brodsky,et al.  Efficacy and Safety of Oral Dofetilide in Converting to and Maintaining Sinus Rhythm in Patients With Chronic Atrial Fibrillation or Atrial Flutter: The Symptomatic Atrial Fibrillation Investigative Research on Dofetilide (SAFIRE-D) Study , 2000, Circulation.

[19]  D. Tester,et al.  Diagnostic Miscues in Congenital Long-QT Syndrome , 2007, Circulation.

[20]  R M Sabbatini,et al.  One-year mortality prognosis in heart failure: a neural network approach based on echocardiographic data. , 1995, Journal of the American College of Cardiology.

[21]  M. Santini,et al.  Comparison of intravenously administered dofetilide versus amiodarone in the acute termination of atrial fibrillation and flutter. A multicentre, randomized, double-blind, placebo-controlled study. , 2000, European heart journal.

[22]  S. Adabag,et al.  Discontinuation of Dofetilide From QT Prolongation and Ventricular Tachycardia in the Real World. , 2016, JACC. Clinical electrophysiology.

[23]  Jeffrey M. Hausdorff,et al.  Physionet: Components of a New Research Resource for Complex Physiologic Signals". Circu-lation Vol , 2000 .

[24]  C. Krittanawong,et al.  Artificial Intelligence in Precision Cardiovascular Medicine. , 2017, Journal of the American College of Cardiology.

[25]  J Vicente,et al.  Differentiating Drug‐Induced Multichannel Block on the Electrocardiogram: Randomized Study of Dofetilide, Quinidine, Ranolazine, and Verapamil , 2014, Clinical pharmacology and therapeutics.

[26]  S. Oliver,et al.  The pharmacokinetics and pharmacodynamics of oral dofetilide after twice daily and three times daily dosing. , 2000, British journal of clinical pharmacology.

[27]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[28]  D. Jonker,et al.  A Pharmacokinetic‐pharmacodynamic Model for the Quantitative Prediction of Dofetilide Clinical QT Prolongation from Human Ether‐a‐go‐go‐related Gene Current Inhibition Data , 2005, Clinical pharmacology and therapeutics.

[29]  J. Dudley,et al.  Cognitive Machine-Learning Algorithm for Cardiac ImagingCLINICAL PERSPECTIVE , 2016 .

[30]  L. Køber,et al.  Efficacy of Dofetilide in the Treatment of Atrial Fibrillation-Flutter in Patients With Reduced Left Ventricular Function: A Danish Investigations of Arrhythmia and Mortality ON Dofetilide (DIAMOND) Substudy , 2001, Circulation.

[31]  Li Zhang,et al.  Inaccurate electrocardiographic interpretation of long QT: the majority of physicians cannot recognize a long QT when they see one. , 2005, Heart rhythm.

[32]  Arthur A M Wilde,et al.  Accurate electrocardiographic assessment of the QT interval: teach the tangent. , 2008, Heart rhythm.

[33]  Fei Wang,et al.  Deep learning for healthcare: review, opportunities and challenges , 2018, Briefings Bioinform..

[34]  Andrew Y. Ng,et al.  Cardiologist-Level Arrhythmia Detection with Convolutional Neural Networks , 2017, ArXiv.

[35]  P. Macfarlane,et al.  Comparison of automated measurements of electrocardiographic intervals and durations by computer-based algorithms of digital electrocardiographs. , 2014, American heart journal.

[36]  M. Chung,et al.  Safety of Oral Dofetilide Reloading for Treatment of Atrial Arrhythmias , 2017, Circulation: Arrhythmia and Electrophysiology.

[37]  Dulciana D. Chan,et al.  Late sodium current block for drug‐induced long QT syndrome: Results from a prospective clinical trial , 2016, Clinical pharmacology and therapeutics.

[38]  J. Dudley,et al.  Machine-Learning Algorithms to Automate Morphological and Functional Assessments in 2D Echocardiography. , 2016, Journal of the American College of Cardiology.

[39]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[40]  S. Cobbe,et al.  Effects of the class III antiarrhythmic drug dofetilide on ventricular monophasic action potential duration and QT interval dispersion in stable angina pectoris. , 1992, The American journal of cardiology.

[41]  T. Lancet.  Artificial intelligence in health care: within touching distance , 2018, The Lancet.

[42]  A Murray,et al.  Accuracy of four automatic QT measurement techniques in cardiac patients and healthy subjects. , 1996, Heart.