Analysis of Image Featuers for Efficient Classification of Thyroid Diseases


  • S. Premalatha
  • S. Subha


In current scenario thyroid disorders has become common among the general public/ population. The thyroid gland is positioned in the anterior part of the neck region between the vertebrae T1 and C5 and the lobes of this gland are held together by Isthmus. Computer based system a technique is used for intelligent categorization of these diseases and it is very useful in diseases and diagnostics management. The purpose of the study was to classify the various thyroid types by quantifying the histological image features of Ultrasound images. Six image features were characterized in thyroid tissues, which includes, entropy, mean, brightness, hue, energy and standard deviation oh brightness. To analysis the features statistical stepwise selection and multiple discriminate analyses were used. A new approach which classifies these thyroid diseases is done on the basis of segmentation method. To sort these kinds of diseases some of the techniques used are image enhancement processing to avoid noise and gray level compensation for segmentation.