Fetal Heart Disease Detection Via Deep Reg Network Based on Ultrasound Images DOI Creative Commons
S. Magesh,

P.S. Raja Kumar

Journal of Applied Engineering and Technological Science (JAETS), Год журнала: 2023, Номер 5(1), С. 439 - 450

Опубликована: Дек. 10, 2023

Congenital heart disease (CHD) is the most prevalent congenital ailment. One in every four newborns born with serious coronary artery will require surgery or other early therapy. Early identification of CHD fetal heart, on hand, more critical for diagnosis. Extracting information from ultrasound (US) images a difficult and time-consuming job. Deep learning (Dl) CNNs have been frequently utilized echocardiography CAD to overcome this difficulty. In work, DL based neural network proposed classifying normal abnormal US images. A total 363 pregnant women between ages 18 34 weeks who had good hearts were included. These are pre-processed using SCRAB (scalable range adaptive bilateral filter) eliminating noise artifacts. The relevant features extracted classify them into by deep Reg net network. model integrates -module CNN architecture diminish computational complexity and, simultaneously, attains an effectual classification accuracy. higher accuracy 98.4% 97.2% CHD. To confirm efficiency compared various networks.

Язык: Английский

Fetal Heart Disease Detection Via Deep Reg Network Based on Ultrasound Images DOI Creative Commons
S. Magesh,

P.S. Raja Kumar

Journal of Applied Engineering and Technological Science (JAETS), Год журнала: 2023, Номер 5(1), С. 439 - 450

Опубликована: Дек. 10, 2023

Congenital heart disease (CHD) is the most prevalent congenital ailment. One in every four newborns born with serious coronary artery will require surgery or other early therapy. Early identification of CHD fetal heart, on hand, more critical for diagnosis. Extracting information from ultrasound (US) images a difficult and time-consuming job. Deep learning (Dl) CNNs have been frequently utilized echocardiography CAD to overcome this difficulty. In work, DL based neural network proposed classifying normal abnormal US images. A total 363 pregnant women between ages 18 34 weeks who had good hearts were included. These are pre-processed using SCRAB (scalable range adaptive bilateral filter) eliminating noise artifacts. The relevant features extracted classify them into by deep Reg net network. model integrates -module CNN architecture diminish computational complexity and, simultaneously, attains an effectual classification accuracy. higher accuracy 98.4% 97.2% CHD. To confirm efficiency compared various networks.

Язык: Английский

Процитировано

2