Medical diagnosis of thyroid disorders using ultrasound images

Abstract

iii newlineAbstract Today, diagnosis of a patient is rarely done without the use of imaging technology. In the field of health care, many automated systems are implemented for diagnosis of numerous diseases using medical images. Several researchers are using statistical, machine learning and image processing techniques to develop automated systems which help health care professionals in the diagnosis of various diseases like Diabetes, Stroke, Lung cancer, Breast Cancer, Hepatitis, Heart Disease, Tuberculosis etc. The automated systems implemented using image processing and machine learning techniques offer doctors with high resolution images which have greatly assisted the clinical diagnosis. These automated systems have a dynamic role to play, as the ratio of doctors to patients is poor. The present research work is set with the intent of developing a computerized system by applying innovative techniques of machine learning, deep learning, image processing concurrently aided with soft computing techniques in the area of Thyroid disorders using ultrasound images. Thyroid disorders denote an important subset of endocrine disorders which are most common among Indian population. Thyroid disorders are the second most common in endocrine pathology. Worldwide approximately 8-20% of adult population are affected with thyroid-related diseases, out of which, less than 5% are diagnosed with cancer. The burden of thyroid disorders in the general population is enormous. The most common cause of thyroid disorders is a primary failure of the thyroid gland. Recent reports show that 300 million people in the world are suffering from thyroid disorders and among them about 42 million people resides in India. Thyroid disease diagnosis using image processing techniques involves image pre-processing, ROI definition, image segmentation, feature extraction, feature selection, classification and post-processing. Systems developed using these innovative methods will assist doctors in early diagnosis of different Thyroid disorders thus sav