Anemia Detection through Sclera and Vessel Analysis: A Machine Learning and Deep Learning Perspective

Anemia needs to be properly and promptly diagnosed because it is a major worldwide health concern. A decrease inred blood cells is known as anemia. In this research, we have used a range of machine learning techniques, including Random Forest, k-Nearest Neighbours, Polynomial Support Vector Machines, and AdaBoost. Specifically, these techniques have been applied to identify patterns and correlations within the extensive information on the diagnosis of anemia. In this work, incorporating these numerous models into an ensemble has been a significant and innovative tactic. This ensemble model, which combines the best aspects of Random Forest, Polynomial SVM, AdaBoost, and k-Nearest Neighbours, has produced remarkable results. It has been noted for its strong performance in the vessel and sclera classifications. We studied machine learning techniques before moving on to advanced computational techniques, such as deep learning approaches. We have primarily discussed convolutional neural networks, covering well-known models such as VGG16, VGG19, MobileNetV2, and InceptionV3. A detailed analysis of CNN-based models has yielded several surprising conclusions: We have observed very high accuracy rates for the vascular and sclera categories, respectively. This study has demonstrated the significant impact of state-of-the-art computational technologies on the delivery of healthcare solutions. Recent works have extended the reach of anemia detection capabilities and added to the increasing corpus of literature, highlighting the critical role that technology plays in managing complex medical scenarios.