Key result
A ResNet-101 classifier using polysomnography and physiological data achieved 94.87% accuracy in classifying cardiovascular disease risk among patients with obstructive sleep apnea.
Why the study?
Obstructive sleep apnea is linked to cardiovascular disease risk, prompting the development of models to classify CVD risk in patients with OSA.
Can a ResNet-101 classifier accurately predict cardiovascular disease risk in patients with obstructive sleep apnea using polysomnography and physiological data?
Can a ResNet-101 classifier accurately predict cardiovascular disease risk in patients with obstructive sleep apnea using polysomnography and physiological data?
A deep learning model using polysomnography and physiological data can accurately classify cardiovascular disease risk in patients with obstructive sleep apnea.
May support deep learning CVD risk classification in OSA; leaves open need for prospective validation before clinical adoption.
Cardiovascular disease (CVD) is a disorder that affects the heart function and blood vessels, such as coronary heart disease, arrhythmias, heart attack, rheumatic heart disease, stroke, etc. The risks associated with CVD are physical inactivity, improper diet, and smoking and liquor drinking. Some behavioral risk factors include increased blood sugar and blood pressure levels, obesity, stress, and genetic factors. Obstructive sleep apnea (OSA) is a disorder related to sleep respiration that affects blood oxygen concentration and air entering the lungs. It will probably lead to overweight, heart disease, diabetes, stroke, and hypertension. Several kinds of research are being carried out, and they demonstrate the relationship between patients with CVD and those with OSA. The present study uses polysomnography data from the Sleep Heart Health Study and CVD-related physiological data obtained from the National Sleep Research Resource. The data set is preprocessed, and principal component analysis reduction is applied and fed to the ResNet-101 classifier for classifying patients with OSA into three types of coronary heart disease (CHD) is a disorder in which the coronary arteries were blocked and CVD risk groups. After training the proposed network with 80 % of records, accuracy was obtained as 94.87 % for “no risk” and “risk,” with a 95.89 % level of CVD risk among patients with OSA apart from its capability to detect the severity of OSA.
No takes yet. Share an insight, caveat, or question.
Priyadharshini et al. (2023) studied Obstructive sleep apnea and cardiovascular disease risk. ResNet-101 classifier with principal component analysis was evaluated on Classification accuracy for 'no risk' and 'risk' of cardiovascular disease. A ResNet-101 classifier using polysomnography and physiological data achieved 94.87% accuracy in classifying cardiovascular disease risk among patients with obstructive sleep apnea.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: