Title : AI based detection of iron deficiency anemia using conjunctiva of images
Abstract:
Anemia is an iron deficiency that often results in the decrease of Red Blood Cells (RBC). One of the worldwide public health issues affecting children and pregnant women is anemia. The main cause of anemia is increased RBC destruction, blood loss, and defective cell production. It also occurs when the amount of RBC within the body decreases, or when the Hemoglobin level of blood is below the normal threshold. Achieving a high level of accurate detection of anemia is a challenging task in existing models. Anxiety and expense are two difficulties in the invasive method of identifying anemia, which prevents advancements in health. Hence, it is essential to develop non-invasive methods for diagnosing anemia that reduce expenses and increase detection efficacy. This proposed model helps to detect iron deficiency anemia from the conjunctiva images of humans using the deep learning model. The early detection of iron deficiency anemia is helpful to take proper medical facilities to enhance the quality of life of the human. The conjunctiva images required to perform the anemia detection are collected from the benchmark dataset. The collected conjunctiva images are directly passed to the Nested Dilated Efficient Attention Network (NDEAN) for detecting iron deficiency in humans. The developed model is used to find the anemic state of the human in a more efficient way. Finally, the effectiveness of the NDEAN is determined via the validation process among several performance indices.
Keywords: Iron Deficiency Anemia Detection; Conjunctiva Images; NDEAN-Nested Dilated Efficient Attention Network.
