Predicting Blood Hemoglobin Level With Machine Learning And Spectroscopy

Anemia, a widespread health condition characterized by low hemoglobin or red blood cell levels, remains a significant global concern, especially in resource-limited settings. The current methods for diagnosing anemia often require invasive blood tests, which can be costly and inaccessible for routine monitoring. This study presents a Python-based model aimed at providing a non-invasive, automated method for anemia detection by analyzing the conjunctival region of the eye. Utilizing Haar cascade classifiers for precise eye region detection and OpenCV for image processing, the model efficiently extracts and analyzes the RGB values of the conjunctiva. These values are compared to established thresholds indicative of hemoglobin deficiency, allowing for a preliminary diagnosis. The model, trained and tested with real-world image data, offers a promising solution to early anemia detection, combining rapid processing, cost-effectiveness, and accessibility. The strength of this work lies in its automation, affordability, and potential for application in remote or underserved areas where traditional diagnostic methods are not readily available.