Medical images segmentation using gabor filters applied to echocardiographic images

Visualisation of medical images, and the subsequent analysis by a specialised physician, always conveys a degree of uncertainty when identifying an illness or condition, as in the case of echocardiographic images. Texture segmentation involves the identification of uniform regions within a given image. The Multichannel Filtering theory is based on the following hypothesis: human visual system in its early stages discomposes the retina into a large set of filtered images, where each one contains the intensity variations in a narrow frequency range at a given orientation. This work presents a texture segmentation algorithm based on the Multi-channel Filtering theory applied on echocardiographic images in order to detect a thrombus; each channel has been obtained using a Gabor filter uniformly distributed along all of the space-frequency domain. Using this procedure on test images and echocardiographic images, we obtained good results for segmentation.