Teaching automated diagnostic systems for Doppler ultrasound blood flow signals to biomedical engineering students using MATLAB

Abstract

This paper presents an initiative to teach the concept of automated diagnostic systems for Doppler ultrasound blood flow signals to biomedical engineering students. The approach was based on illustrative applications that highlight the performance of multilayer perceptron neural networks (MLPNN) and adaptive neuro-jazzy inference system (ANFIS). Following a brief description of the artificial neural networks (ANNs) and ANFIS, applications of the models to the Doppler signals ohtained from ophthalmic artery and internal carotid artery were done by means of a series of MATLAB functions. The functions involved in the neural network and fuzzy logic toolboxes of MATLAB can be used to develop automated diagnostic systems for the signal under study. The authors suggest that the use of MATLAB exercises will assist the students in gaining a better understanding of the various automated diagnostic systems in blood flow signals.