Knowledge Base System for Diagnostic Assessment of Doppler Spectogram (original) (raw)

2000

Abstract

The aim of this paper is to develop an automated diagnosis system for arterial diseases from the Doppler ultrasonography spectrogram. A feature-based classification of Doppler spectrogram has been done for its clinical assessment. Development of the automated diagnostic tool for Doppler spectrogram involves three steps (a) feature extraction, (b) classification of spectrogram based on clinical symptoms and (c) diagnosis of arterial condition of the concerned region. Artificial Neural Network is used for classification of the spectrograms. Arterial condition of a specified region is evaluated from symptoms obtained from a number of spectrograms in the different nearby regions. Bayesian probabilistic method is used for diagnostic evaluation of the arterial status from the obtained spectrogram. The results satisfied 83% of the obtained cases.

Bipul Das hasn't uploaded this paper.

Let Bipul know you want this paper to be uploaded.

Ask for this paper to be uploaded.