A Low-Cost Test for Anemia Using an Artificial Neural Network.

Area of Science:

• Medical Diagnostics
• Artificial Intelligence in Healthcare
• Biomedical Imaging

Background:

• Anemia in pregnancy poses significant risks to maternal and neonatal health, particularly in low-income countries.
• Early anemia screening is crucial for improving pregnancy outcomes.
• Point-of-care techniques are vital for rapid diagnosis in resource-constrained settings.

Purpose of the Study:

• To develop an Artificial Neural Network (ANN) tool for estimating hemoglobin levels.
• To utilize color information from blood sample images for hemoglobin estimation.
• To create an inexpensive and easily assembled image acquisition setup.

Main Methods:

• Collected blood samples from 86 volunteers.
• Acquired images of blood drops using a custom-designed setup.
• Used color intensity values from images as features for an Artificial Neural Network model.

Main Results:

• The best ANN model estimated hemoglobin with an accuracy of ±1.8 g/dl LOA and 0.03 g/dl bias.
• Validation with 65 additional samples showed ±2 g/dl to -1.9 g/dl LOA and 0.06 g/dl bias.
• Achieved 95.5% sensitivity and 52% specificity.

Conclusions:

• The developed ANN method provides accurate hemoglobin estimations.
• The technique is comparable to contemporary measurement methods.
• This approach offers a viable screening technique for anemia, especially in resource-limited areas.