Raspberry Pi-Based Non-Invasive Anemia Detection via Palpebral Conjunctiva Image Analysis

This study proposes a low-cost, non-invasive anemia detection system based on palpebral conjunctiva image analysis using a Raspberry Pi platform. The system captures high-quality images of the palpebral conjunctiva—the inner surface of the lower eyelid—using a USB webcam connected to the Raspberry Pi. These images are pre-processed to enhance relevant features, such as color and texture, which are indicative of hemoglobin levels. A dataset of labeled palpebral conjunctiva images, collected from subjects with varying degrees of anemia, is used to train machine learning models. Feature extraction techniques, including color histogram analysis and image segmentation, are applied to highlight the conjunctiva region. The extracted features are then fed into a classifier, such as Support Vector Machine (SVM) or a lightweight Convolutional Neural Network (CNN), which is trained to distinguish between normal and anemic cases, as well as classify the severity of anemia. The trained model is deployed on the Raspberry Pi, enabling real-time anemia detection and classification through image analysis displayed on an attached LCD screen. This approach offers a portable, accessible, and efficient tool for anemia screening, especially useful in remote or resource-limited areas where conventional blood testing is not feasible.