Using Deep Neural Network to Detect Thyroid Cancer

Abstract

Because of its butterfly-like structure, the thyroid gland is also referred to as the butterfly gland. This gland is located in the neck and controls metabolism. It is responsible for the synthesis of the hormones triiodothyronine (T3) and thyroxin (T4), both of which operate as chemical “messengers” by travelling via the bloodstream to every single cells and tissues in the body. These hormones are produced by the thyroid gland. This helps to keep our “metabolism,” which is another word for the rate at which our body functions, at a consistent level. There are some people whose thyroid glands do not operate as effectively as they should. Hypothyroidism is a disorder that happens when the thyroid gland releases insufficient thyroid hormone, which can be induced by an underactive thyroid gland. An underactive thyroid gland can also contribute to the development of hypothyroidism. Some people have an overactive thyroid gland, which can lead to hyperthyroidism, which is characterized by excessively high amounts of the thyroid hormone. Both of these conditions are considered to be major kinds of thyroid illness. A slowing of the body's metabolic rate is one of the symptoms of hypothyroidism. Gaining weight, feeling tired all the time, having problems remembering things, having trouble moving and thinking, having sluggish speech, having problems concentrating, and anxiety are all possible symptoms. On the other hand, hyperthyroidism quickens the rate at which the body burns fuel (its metabolism). This results in a wide range of symptoms, such as a rapid heartbeat, decreased appetite, diarrhoea, thirst, feeling shaky and sweaty, experiencing an abnormally high body temperature, and so on. There is a plethora of different approaches that can be utilized in order to solve this problem, some of which include machine learning (ML) and DL approaches. The primary objective of this piece of research is to develop a novel accuracy that the researchers who make use of deep NN would find useful.