A deep neural network electrocardiogram analysis framework for left ventricular hypertrophy prediction

2020 ◽  
Author(s):  
Jimmy Ming-Tai Wu ◽  
Meng-Hsiun Tsai ◽  
Sheng-Han Xiao ◽  
Yung-Po Liaw
Keyword(s):  
Neural Network ◽  
Left Ventricular Hypertrophy ◽  
Ventricular Hypertrophy ◽  
Deep Neural Network ◽  
Left Ventricular ◽  
Analysis Framework

Comparing the performance of artificial intelligence and conventional diagnosis criteria for detecting left ventricular hypertrophy using electrocardiography

EP Europace ◽  
2019 ◽  
Vol 22 (3) ◽  
pp. 412-419 ◽  
Author(s):  
Joon-Myoung Kwon ◽  
Ki-Hyun Jeon ◽  
Hyue Mee Kim ◽  
Min Jeong Kim ◽  
Sung Min Lim ◽  
...  
Keyword(s):  
Neural Network ◽  
Artificial Intelligence ◽  
Left Ventricular Hypertrophy ◽  
Ventricular Hypertrophy ◽  
External Validation ◽  
Left Ventricular ◽  
Machine Learning Techniques ◽  
Validation Dataset ◽  
Point System ◽  
Diagnosis Criteria

Abstract Aims  Although left ventricular hypertrophy (LVH) has a high incidence and clinical importance, the conventional diagnosis criteria for detecting LVH using electrocardiography (ECG) has not been satisfied. We aimed to develop an artificial intelligence (AI) algorithm for detecting LVH. Methods and results This retrospective cohort study involved the review of 21 286 patients who were admitted to two hospitals between October 2016 and July 2018 and underwent 12-lead ECG and echocardiography within 4 weeks. The patients in one hospital were divided into a derivation and internal validation dataset, while the patients in the other hospital were included in only an external validation dataset. An AI algorithm based on an ensemble neural network (ENN) combining convolutional and deep neural network was developed using the derivation dataset. And we visualized the ECG area that the AI algorithm used to make the decision. The area under the receiver operating characteristic curve of the AI algorithm based on ENN was 0.880 (95% confidence interval 0.877–0.883) and 0.868 (0.865–0.871) during the internal and external validations. These results significantly outperformed the cardiologist’s clinical assessment with Romhilt-Estes point system and Cornell voltage criteria, Sokolov-Lyon criteria, and interpretation of ECG machine. At the same specificity, the AI algorithm based on ENN achieved 159.9%, 177.7%, and 143.8% higher sensitivities than those of the cardiologist’s assessment, Sokolov-Lyon criteria, and interpretation of ECG machine. Conclusion  An AI algorithm based on ENN was highly able to detect LVH and outperformed cardiologists, conventional methods, and other machine learning techniques.

Left Ventricular Hypertrophy

2014 ◽  
Vol 19 (2) ◽  
pp. 11-15
Author(s):  
Steven L. Demeter
Keyword(s):  
Risk Factors ◽  
Left Ventricular Hypertrophy ◽  
Ventricular Hypertrophy ◽  
Left Ventricular Mass Index ◽  
Medical Science ◽  
Careful Analysis ◽  
Left Ventricular ◽  
Hypertensive Disease ◽  
Accurate Analysis ◽  
Clear Explanation

Abstract The fourth, fifth, and sixth editions of the AMA Guides to the Evaluation of Permanent Impairment (AMA Guides) use left ventricular hypertrophy (LVH) as a variable to determine impairment caused by hypertensive disease. The issue of LVH, as assessed echocardiographically, is a prime example of medical science being at odds with legal jurisprudence. Some legislatures have allowed any cause of LVH in a hypertensive individual to be an allowed manifestation of hypertensive changes. This situation has arisen because a physician can never say that no component of LVH was not caused by the hypertension, even in an individual with a cardiomyopathy or valvular disorder. This article recommends that evaluators consider three points: if the cause of the LVH is hypertension, is the examinee at maximum medical improvement; is the LVH caused by hypertension or another factor; and, if apportionment is allowed, then a careful analysis of the risk factors for other disorders associated with LVH is necessary. The left ventricular mass index should be present in the echocardiogram report and can guide the interpretation of the alleged LVH; if not present, it should be requested because it facilitates a more accurate analysis. Further, if the cause of the LVH is more likely independent of the hypertension, then careful reasoning and an explanation should be included in the impairment report. If hypertension is only a partial cause, a reasoned analysis and clear explanation of the apportionment are required.

The association of ankle brachial index with left ventricular hypertrophy and left ventricular mass index: the Dong-gu study

VASA ◽  
2013 ◽  
Vol 42 (4) ◽  
pp. 284-291 ◽  
Author(s):  
Seong-Woo Choi ◽  
Hye-Yeon Kim ◽  
Hye-Ran Ahn ◽  
Young-Hoon Lee ◽  
Sun-Seog Kweon ◽  
...  
Keyword(s):  
Left Ventricular Hypertrophy ◽  
Left Ventricular Mass ◽  
Ventricular Hypertrophy ◽  
Left Ventricular Mass Index ◽  
Ankle Brachial Index ◽  
Intima Media Thickness ◽  
Left Ventricular ◽  
Baseline Survey ◽  
Ventricular Mass Index ◽  
Ventricular Mass

Background: To investigate the association between ankle-brachial index (ABI), left ventricular hypertrophy (LVH) and left ventricular mass index (LVMI) in a general population. Patients and methods: The study population consisted of 8,246 people aged 50 years and older who participated in the baseline survey of the Dong-gu Study conducted in Korea between 2007 and 2010. Trained research technicians measured LV mass using mode M ultrasound echocardiography and ABI using an oscillometric method. Results: After adjustment for risk factors and common carotid artery intima-media thickness (CCA-IMT) and the number of plaques, higher ABIs (1.10 1.19, 1.20 - 1.29, and ≥ 1.30) were significantly and linearly associated with high LVMI (1.10 - 1.19 ABI: β, 3.33; 95 % CI, 1.72 - 4.93; 1.20 - 1.29 ABI: β, 6.51; 95 % CI, 4.02 - 9.00; ≥ 1.30 ABI: β, 14.83; 95 % CI, 6.18 - 23.48). An ABI of 1.10 - 1.19 and 1.20 - 1.29 ABI was significantly associated with LVH (1.10 - 1.19 ABI: OR, 1.35; 95 % CI, 1.19 - 1.53; 1.20 - 1.29 ABI: OR, 1.59; 95 % CI, 1.31 - 1.92) and ABI ≥ 1.30 was marginally associated with LVH (OR, 1.73; 95 % CI, 0.93 - 3.22, p = 0.078). Conclusions: After adjustment for other cardiovascular variables and CCA-IMT and the number of plaques, higher ABIs are associated with LVH and LVMI in Koreans aged 50 years and older.

Sign in / Sign up

Export Citation Format

Share Document