Pituitary Gland Size Estimation and Lesion Detection using Segmentation-based Color Thresholding Technique

The pituitary gland pea-sized gland situated in the human brain below hypothalamus. This is responsible for generating useful hormones in the human body. These hormones are used to carry the message from blood to organs. The size of the gland is very important. Hence, this study presents a novel image segmentation method that can be used to identify and quantify the percentage of pituitary glands in sagittal Magnetic Resonance Imaging (MRI) images. The proposed method utilizes a color thresholding technique in MATLAB software to separate the pituitary gland from the surrounding tissues. The color thresholding technique involves converting the original RGB images to HSV color space, which is a cylindrical representation of colors using three components: hue, saturation, and value. The presence and absence of glands can be analyzed by histogram analysis. It can be used to define threshold values as binary masks. The binary mask can then be used to calculate the percentage of the pituitary gland in the image.