Computerized segmentation of sinus images

Sinusitis is diagnosed with techniques such as endoscopy, ultrasound, X-ray, computed tomography (CT) scan and magnetic resonance imaging (MRI). Out of these techniques, imaging techniques are less invasive while being able to show blockage of sinus cavities. However, the potential of these techniques have not been fully realised as the images obtained are still bound to misinterpretations. This project attempts to solve this problem by developing an algorithm for the computerized segmentation of sinus images for the detection of sinusitis. The image enhancement techniques used were median filtering and the contrast limited adapted histogram equalisation (CLAHE) method. These techniques applied on input images managed to reduce noise and smoothen the image histogram. Multilevel thresholding algorithms were developed to segment the images into meaningful regions for the detection and diagnosis of sinusitis. These algorithms were able to extract important features from the images. The software used for simulations is Matlab. Simulations were performed on images of healthy sinuses and sinuses with sinusitis. The algorithms were able to differentiate between healthy sinuses and sinuses with sinusitis.