Deep Convolutional Neural Network Model for Concurrent Prediction of Anemia and Blood Glucose Level from Conjunctiva Images

Evaluation of blood glucose levels and anemia are important components of healthcare that influence early diagnosis and therapy. This article provides a unique method based on conjunctiva images that uses deep convolutional neural networks (CNNs) to predict blood glucose levels and anemia status simultaneously. The suggested model shows remarkable accuracy and a high success rate in differentiating between anemic and non-anemic people. Furthermore, it provides an average mean squared error for accurate blood glucose estimation. We confirm the model's dependability as a non-invasive option by comparing its performance to conventional invasive blood tests. The model has the potential to be widely applied in the healthcare industry due to its robustness and generalizability, even when applied to different datasets. In comparison to current practices, this novel technology provides a number of important benefits that improve patient care and diagnostics while reducing pain and annoyance.