Novel hybrid approach based on combination of textural features and clinical parameters for reliable prediction of thyroid malignancy

The dramatic increase in thyroid cancer, particularly among the younger population demands development of an automated decision support system for timely and reliable prognosis disease so as to facilitate improved chances recovery subjects. While numerous methods are already reported by researchers detection Thyroid malignancy, most crucial parameter Malignancy Index (TMI) has received very less attention. TMI is paramount importance diagnosis treatment patients having malignant its consideration therefore inevitable. This research aims develop a malignancy. proposed hybrid approach incorporates novel combination texture features clinically observable parameters initially identify tumor using vector machines further predicts value, thus exhibiting performance like trained radiologist. Publically available database comprising 99 cases 134 ultrasound images used validate superiority approach. Apt prediction this makes designed marks mega leap towards practical deployment clinical environment.