An Identification and Classification of Thyroid Diseases Using Deep Learning Methodology

Abstract

The thyroid is one of the most important parts of our body. As part of the endocrine system, this tiny gland in our neck releases thyroid hormone, which is responsible for directing all your metabolic functions which means controlling everything from digestion to conversion to energy. When thyroid dysfunction, it can affect all aspects of our health. Both researchers and doctors face challenges in fighting thyroid disease. In that thyroid disease is a major cause of the emergence of medical diagnostics and prognosis, the beginning of which is a difficult confirmation in medical research. Thyroid hormones are suspected to regulate metabolism. Hyperthyroidism and hypothyroidism are one of the two most common thyroid diseases that release thyroid hormones to regulate the rate of digestion. Early detection of thyroid disease is a major factor in saving many lives. Frequently, visual tests and hand techniques are used for these types of diagnostic thyroid diseases. This manual interpretation of medical images requires the use of time and is highly affected by errors. This work is developed to successfully diagnose and detect the presence of five different thyroid diseases such as Hyperthyroidism, Hypothyroidism, Thyroid cancer, thyroid gland, Thyroiditis and general thyroid screening without the need for several consultations. This leads to predictable disease progression and allows us to take immediate steps to avoid further consequences in an effective and cost-effective way to avoid the human error rate. A web application will also be developed where a scanned image of the inclusion will provide the removal of the most time-consuming thyroid type and patient investment.

Gait Based Person Identification Using Deep Learning Model of Generative Adversarial Network

“…Several template-based methods [10][11][12] stand out as the most common automatic recognition methods nowadays. On the other hand, a different class of methods represented pedestrians' structural and kinetic properties using coarse human-model techniques [13][14][15]. Over a hundred people and various critical aspects, including view variants and appearance-changing, were considered in the early stages of dataset building the encouraging consequences of the test, it appears that gait recognition is doable and could be useful in the future [16].…”