Computer Simulation And Correction Of Electrocardiogram Voltage To Assist In The Diagnosis Of Cardiac Amyloidosi

Background:Cardiac amyloidosis is a kind of disease in which misfolded proteins are deposited in the myocardium,resulting in impaired diastolic and systolic function of the heart.Although there are as many as 30 types of amyloid protein that can be deposited in the myocardium,immune light chain and transthyretin are two of the most common.Although drugs of significant benefit for both diseases have been recommended in guidelines,delayed treatment due to misdiagnosis can reduce expected survival.The first diagnosis of such patients is often in a primary hospital without technique of myocardial biopsy and radionuclide.Therefore,a simple and easy-to-operate method based on basic examinations with high sensitivity and specificity is needed to screen the disease,and to differentiate it from other diseases that also manifest left ventricular hypertrophy on ultrasound.ECG voltage has long been used to indicate cardiac amyloidosis.However,because the ECG is widely affected by the geometry and orientation of the body and the heart,its low sensitivity has been criticized.If these geometric factors can be properly corrected,it is expected to improve the power of screening and identification of cardiac amyloidosis by ECG.Methods:This study was divided into two parts:computer modeling and clinical cohort validation.In the computer modeling part,the parameters extracted from the medical images are used as the main input,and the theoretical ECG voltage is used as the main output.The main input includes the patient’s left and right thoracic diameter,thoracic anteroposterior diameter,the left and right position of the heart in the thorax,the anterior and posterior position of the heart in the thorax,the left ventricular enddiastolic diameter of the heart,the thickness of the left ventricular posterior wall and the interventricular septum.Novel parametric modeling algorithm was used for fast three-dimensional modeling and meshing.Myocardial electrophysiological model was used for simulation to obtain theoretical ECG voltage and compare with actual ECG voltage to obtain ratio,which is used as diagnosis indicator(ECG Voltage Index).In the clinical validation part,a total of 315 patients in 4 groups of cardiac amyloidosis,hypertrophic cardiomyopathy,hypertensive cardiomyopathy and healthy controls were included.Taking discharge diagnosis as the gold standard,the traditional ECG voltage and the proposed ECG voltage index were used to perform diagnostic tests in each group,respectively,and the diagnostic efficacy of the new and old indexes was compared using the area under the ROC curve.Results:Among the four groups,the average ECG voltage index of the cardiac amyloidosis group was 0.55,which was significantly lower than that of the other three groups(1.4,1.23,and 1.44,respectively,P<0.001).Compared with the traditional Sokolow-Lyon voltage and two commonly used voltage-to-mass ratio indices,the newly proposed ECG voltage index has the largest area under the ROC curve,which can differentiate cardiac amyloidosis from healthy people,hypertrophic cardiomyopathy,and hypertrophic cardiomyopathy.The areas under the ROC curve were 0.943,0.916 and 0.831,respectively.It indicated that the ECG voltage index obtained after simulation-aided correction of thoracic and cardiac geometric parameters could not only distinguish patients from healthy people better,but also better differentiate amyloid heart disease from hypertrophic and hypertensive cardiomyopathy.Conclusion:Computer simulation can effectively correct ECG voltage and improve the diagnosis and screening efficiency of ECG voltage for amyloid cardiomyopathy.