Neural network analysis of Doppler ultrasound blood flow signals: A pilot study

Abstract

It has been hypothesised that each artery in the human body has its own characteristic “signature”—a unique Doppler flow profile which can identify the artery and which may also be modified by the presence of disease. To test this hypothesis an artificial neural network (ANN) was trained to recognise three groups of maximum frequency envelopes derived from Doppler ultrasound spectrograms; these were the common carotid, common femoral and popliteal arteries. Data were collected from 24 subjects known to have no significant atheromatous disease. The maximum frequency envelopes were used to create sets of training and testing vectors for a backpropagation ANN. The ANN demonstrated a high success rate for appropriate classification of the test vectors: 100% for the carotid; 92% for the femoral; and 96% for the popliteal artery. This work has demonstrated the ability of the ANN to differentiate accurately between different and similar flow profiles, outlining the potential of this technology to identify subtle changes induced by the onset of arterial disease within a specific vessel. It should be noted that the ANN not only models the maximum frequency envelope but also, unlike standard indices, makes a decision as to which artery the maximum frequency envelope belongs to, thus providing the potential to obviate human subjective classification.