Objective of this paper: Early detection of anemia (low hemoglobin count) at point-of-care (PoC) in a low resource setting is challenging given the associated capital and recurring costs of the device. In this paper, we have developed a faster and reliable method to screen the hemoglobin level in the whole blood in a resource limited point of care setting. We have also investigated a simple scattering-based technique to predict the overall morphology of the red blood cells. Methods: In this paper, we have developed and validated a digital camera-based spectrometer for the early detection of anemia with the whole blood of 10- $\mu \text{L}$ volume in a low resource PoC setting. The developed device consists of two LEDs [Green (570 nm) and Red (631 nm)], a web camera, a fluid sensing chamber (cuvette holder), and associate electronics. Validation of Beer–Lambert law using the blood of various degree of dilution has been achieved in this in vitro experiment. Major results: A significant number of human subjects (300) having a wide range of hemoglobin counts in a hospital is tested and found to reproduce results from a gold standard automated hematology analyzer. Our developed microcuvette requires only 10- $\mu \text{L}$ unprocessed whole blood sample for the device. Conclusions: The device is expected to serve as a minimally invasive e-health care device for the anemia screening in any resource-limited point of care setting.