Automated Diagnosis and Assessment of Cardiac Structural Alteration in Hypertension Ultrasound Images

Abstract

Hypertension (HTN) is a major risk factor for cardiovascular diseases. At least 45% of deaths due to heart disease and 51% of deaths due to stroke are the result of hypertension. According to research on the prevalence and absolute burden of HTN in India, HTN positively correlated with age and was present in 20.6% of men and 20.9% of women. It was estimated that this trend will increase to 22.9% and 23.6% for men and women, respectively, by 2025. Controlling blood pressure is therefore important to lower both morbidity and mortality. Computer-aided diagnosis (CAD) is a noninvasive technique which can determine subtle myocardial structural changes at an early stage. In this work, we show how a multi-resolution analysis-based CAD system can be utilized for the detection of early HTN-induced left ventricular heart muscle changes with the help of ultrasound imaging. Firstly, features were extracted from the ultrasound imagery, and then the feature dimensions were reduced using a locality sensitive discriminant analysis (LSDA). The decision tree classifier with contourlet and shearlet transform features was later employed for improved performance and maximized accuracy using only two features. The developed model is applicable for the evaluation of cardiac structural alteration in HTN and can be used as a standalone tool in hospitals and polyclinics.

Directional-Guided Motion Sensitive Descriptor for Automated Detection of Hypertension Using Ultrasound Images

“…It is observed from the related work that only two works have been proposed to identify the HTN using fourchamber heart US images [35,36]. Moreover, these works have utilized only a few US images from the patients.…”

“…Moreover, these works have utilized only a few US images from the patients. The sensitivity and specificity achieved by the graph embedding is 100%, however, only one image per subject is utilized [35]. A greater number of frames to cover one complete cardiac cycle is utilized and proposed GWLBP [36].…”

“…In this stage, the dataset was applied to multiple algorithms, and it was found that the MAH algorithm with random forest have attained an accuracy of 99.48%. Apart from questionnaire-based studies, research was also conducted to detect hypertension using various modalities such as ECG [32][33][34], US images [35,36], and Photoplethysmography (PPG) signals [37,38], A machine learning (ML)-based approach was proposed to detect patients with the probability of developing pulmonary hypertension (PH) [39]. The study utilized the data from an electronic health record (EHR) database.…”

“…This raises the need for automated evaluation tools in patients who are diagnosed with HTN which may demonstrate an objective way to assess subtle myocardial changes in the heart. Hence, computer aided diagnostic (CAD) tools have been developed by various researchers as one of the prospective solutions using US images to avoid human errors [35,36,[71][72][73]. US imaging of heart -echocardiography is the widely used imaging modality in the evaluation of heart diseases.…”

Is artificial intelligence the new kid on the block? Sustainable applications in cardiology

“…The role of AI in CV imaging has already been discussed. However, there is an increasing interest in novel and early imaging biomarkers able to predict the presence of CAD and myocardial ischemia [98,99] . In the context of stress echocardiography, ML and DL methods were used to identify regional wall motion abnormalities, showing a great level of reliability [100,101] .…”