Bindushree C, Chinmayee S

DOI: https://doi.org/10.5281/zenodo.23421336

Anemia is a common disease that may occur in individuals due to low levels of hemoglobin in the blood, whereas conventional diagnosis usually requires clinical examination and laboratory tests. Preliminary screening is often hindered by economic, availability, and even early recognition challenges. AnemiaVision is the proposed artificial intelligence system designed to recognize potential cases of anemia from conjunctiva images. AnemiaVision uses a Convolutional Neural Network to process and classify the image data into anemic and non-anemic classes. The model was trained using a dataset of conjunctiva images categorized into Anemic and Non-Anemic classes. The images were preprocessed and normalized before model training. AnemiaVision is comprised of the trained model and a web interface, which allows users to upload a conjunctiva image and receive screening results with model confidence. The system was implemented using Python programming language, TensorFlow/Keras framework, OpenCV computer vision library, and PHP web interface framework. The proposed system was experimentally evaluated, and results showed potential for rapid preliminary screening of anemia. AnemiaVision is proposed as an educational and screening tool, which cannot substitute laboratory tests and clinical diagnosis by medical professionals.

Anemia Detection, Conjunctiva Images, Convolutional Neural Network, Deep Learning, Medical Image Classification, Medical Image Screening, Artificial Intelligence, Image-Based Screening, Computer Vision, TensorFlow, Keras, Healthcare Technology

The suggested anemia detecting system utilizes a CNN model for the initial screening of the disease based on the inputted images of the patient’s conjunctiva. After preprocessing the obtained images, the system uses the convolutional neural network model to extract the relevant data and classify the result into one of the possible categories. The result of the classification is then presented via the AnemiaVision web application. Thus, the system offers a convenient way of initial anemia screening with a professional diagnosis; however, it should be noted that the proposed solution is only a supportive tool that is incapable of replacing the experienced doctor.

1] J. W. Asare et al., “Iron deficiency anemia detection using machine learning models: A comparative study of fingernails, palm and conjunctiva of the eye images,” Engineering Reports, vol. 5, no. 11, 2023.
[2] N. Sehar et al., “Deep Learning Model-Based Detection of Anemia from Conjunctiva Images,” Healthc Inform. Res., vol. 31, no. 1, pp. 57–65, 2025.
[3] G. S. Raju et al., “Anemia Detection by Exploiting Fingernail Attributes Using Deep Learning,” Int. J. Res. Anal. Rev., vol. 11, no. 2, 2024.
[4] M. Bhatt and P. Kumar, “Detection of Iron Deficiency Anemia using Machine Learning Techniques,” 2022.
[5] S. Mythili et al., “Anaemia Detection using Deep Learning by Processing Conjunctiva, Fingernails, Palm and Tongue Images,” 2024.
[6] P. Appiahene et al., “Application of Machine Learning Approach for Iron Deficiency Anaemia Detection in Children using Conjunctiva Images,” 2024.
[7] K. Uma et al., “Analysis of Anemia Detection from Images of Eye Conjunctiva Using Machine Learning Algorithms,” 2025.
[8] TensorFlow Documentation, “EfficientNetB0 Model Architecture,” Available: tensorflow.org/api_docs/python/tf/keras/applications/Efficie ntNetB0.
[9] F. Chollet, Deep Learning with Python. Manning Publications, 2021.
[10] I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. Cambridge, MA: MIT Press, 2016.

Bindushree C, Chinmayee S (2026). AI-Based Early Prediction of Iron-Deficiency Anemia in Rural Women. International Journal of Computer Science Engineering Techniques, 10(5), 321–324. ISSN: 2455-135X. DOI: https://doi.org/10.5281/zenodo.23421336