AbstractAnemia is a condition observed when the red blood cells or hemoglobin in the bloodstream decrease in number. It is the most common blood condition and affects almost a quarter of the world's population. Severe malnutrition, parasitic infections, or underlying diseases are often the causes of severe anemia. It can also pose as a significant risk factor for death and morbidity, especially in vulnerable groups like children, older adults, and those with chronic illnesses. Our system proposes an automated, easy-to-use, non-invasive technique to detect anemia by analyzing the anterior conjunctival pallor of the eye. For this, we have used a deep learning model developed using the DenseNets-121 algorithm. The training and testing datasets consist of images taken from a smartphone camera along with the necessary normalizations. The system provides adequate accuracy and, thus, can be used as a preliminary screening test for anemia.KeywordsAnemia detectionMachine learningNon-invasiveCNNDenseNets-121
CITATION STYLE
Shaju, A., Shah, A., Iyer, G., Pandya, P., & Sawant, V. (2023). Non-Invasive Anemia Detection Using Images Acquired from Smartphone Camera (pp. 803–813). https://doi.org/10.1007/978-981-19-3951-8_61
Mendeley helps you to discover research relevant for your work.