Advancing Hematological Diagnostics in Resource-Constrained Settings: A Robust Deep Learning Solution for Sickle Cell Anemia Screening

Neelkamal Vishwakarma; Gunjan Kumar Mishra; Gaurav Chaubey

doi:10.69968/ijisem.2025v4i4110-118

Authors

Neelkamal Vishwakarma Research Scholar, Department of Electronics & Communication, SHEAT Babatpur, Varanasi
Gunjan Kumar Mishra Assistant Professor, Department of Electronics & Communication, SHEAT Babatpur, Varanasi
Gaurav Chaubey Assistant Professor, Department of Electronics & Communication, SHEAT Babatpur, Varanasi

DOI:

https://doi.org/10.69968/ijisem.2025v4i4110-118

Keywords:

Sickle Cell Anemia, Machine learning, Data integration, Predictive Modeling, Early diagnosis

Abstract

This study introduces a deep learning model for the identification of sickle cell anaemia (SCA) in red blood cell (RBC) images, with the objective of facilitating efficient diagnostics in resource-limited environments. The study used a DenseNet121-based convolutional neural network (CNN) with a custom classification head and a unique dataset of 691 images from Uganda's Teso region that were processed with Field and Leichman stains. The strong method included using RandomUnderSampler and SMOTE to fix class imbalance, as well as Albumentations to add more data to make the model more resistant to image variability, like differences in staining and blurry images. The model was trained with a 4-fold cross-validation method. In just nine epochs, it reached a peak training accuracy of 96.4% and a loss of 4.0153. This better efficiency and performance beat work that used an InceptionV3 model (91% accuracy, 28.7 loss, 100 epochs). The model's high accuracy and robustness suggest that it could be used in clinical settings in areas where SCA is common, which would be a big step forward for automated haematological diagnostics.

References

[1] Pauling, Linus, et al. "Sickle cell anemia, a molecular disease." Science 110.2865 (1949): 543-548.

https://doi.org/10.1126/science.110.2865.543

[2] Tisdale, John F., Swee Lay Thein, and William A. Eaton. "Treating sickle cell anemia." Science 367.6483 (2020): 1198-1199.

https://doi.org/10.1126/science.aba3827

[3] Adio, Waheed Sakariyau, et al. "Current trends in the treatment of sickle cell anemia." World News of Natural Sciences (WNOFNS) 43 (2022).

[4] Buhari, Hauwa Ali, Aisha Sa'ad Ahmad, and Emmanuel Ifeanyi Obeagu. "Current advances in the diagnosis and treatment of sickle cell anaemia." (2023).

[5] Brandow, A. M., and R. I. Liem. "Advances in the diagnosis and treatment of sickle cell disease." Journal of hematology & oncology 15.1 (2022): 20.

https://doi.org/10.1186/s13045-022-01237-z

[6] Simon, Kawuma, et al. "Comparison of deep learning techniques in detection of sickle cell disease." Artificial Intelligence and Applications. Vol. 1. No. 4. 2023.

https://doi.org/10.47852/bonviewAIA3202853

[7] Bushra, S. Nikkath, and G. Shobana. "Paediatric sickle cell detection using deep learning-a review." 2021 International Conference on Artificial Intelligence and Smart Systems (ICAIS). IEEE, 2021.

https://doi.org/10.1109/ICAIS50930.2021.9395756

[8] D. K. Gaikwad, V. Mahale, A. Gaikwad, and A. Gaikwad, "Automatic Blood Disease Identification System Using Machine Learning Algorithms with Hemogram Reports," INDIAN J. Sci. Technol., vol. 18, no. 1, pp. 11-22, 2025.

https://doi.org/10.17485/IJST/v18i1.3443

[9] J. G. Gómez, C. Parra Urueta, D. S. Álvarez, V. Hernández Riaño, and G. Ramirez-Gonzalez, "Anemia Classification System Using Machine Learning," Informatics, vol. 12, no. 1, pp. 1-16, 2025,

https://doi.org/10.3390/informatics12010019

[10] S. K and A. D. K. T. P, "Sickle Cell Anemia Detection using Deep Learning," Int. J. Sci. Res. Technol., vol. 2, no. 1, pp. 15-26, 2025,

https://doi.org/10.21608/ijaiet.2024.240910.1002

[11] A. Deo, I. Pandey, S. S. Khan, A. Mandlik, N. V. Doohan, and B. Panchal, "Deep Learning-Based Red Blood Cell Classification for Sickle Cell Anemia Diagnosis Using Hybrid CNN-LSTM Model," Internetional Inf. Eng. Technol. Assoc., vol. 41, no. 3, pp. 259-302, 2024,

https://doi.org/10.2307/j.ctv18ph9kr.16

[12] N. G. Goswami, A. Goswami, N. Sampathila, M. G. Bairy, K. Chadaga, and S. Belurkar, "Detection of sickle cell disease using deep neural networks and explainable artificial intelligence," J. Intell. Syst., vol. 33, no. 1, 2024,

https://doi.org/10.1515/jisys-2023-0179

[13] N. G. Goswami, N. Sampathila, G. M. Bairy, A. Goswami, D. D. Brp Siddarama, and S. Belurkar, "Explainable Artificial Intelligence and Deep Learning Methods for the Detection of Sickle Cell by Capturing the Digital Images of Blood Smears," Inf., vol. 15, no. 7, 2024,

https://doi.org/10.3390/info15070403

[14] T. F. Machado, F. das Chagas Barros Neto, M. de Souza Gonçalves, C. G. Barbosa, and M. E. Barreto, "Exploring machine learning algorithms in sickle cell disease patient data: A systematic review," PLoS One, vol. 19, no. 11, pp. 1-17, 2024,

https://doi.org/10.1371/journal.pone.0313315

[15] R. Pandey, S. Ahmad, S. Yadav, and S. Srivastava, "Anemia Detection Using Machine Learning and Deep Learning - A Systematic Literature Review," Int. J. Adv. Res. Innov. Ideas Educ., vol. 10, no. 1, pp. 368-377, 2024.

[16] M. Ramzan, J. Sheng, M. U. Saeed, B. Wang, and F. Z. Duraihem, "Revolutionizing anemia detection: integrative machine learning models and advanced attention mechanisms," Vis. Comput. Ind. Biomed. Art, vol. 7, no. 18, 2024,

https://doi.org/10.1186/s42492-024-00169-4

[17] Sachin and A. Bhadrashetty, "MACHINE LEARNING BASED PREDICTION OF SICKLE CELL ANEMIA IN HUMAN RBC," Int. Res. J. Mod. Eng. Technol. Sci., vol. 6, no. 5, 2024.

[18] Sen, Bheem, et al. "Deep Learning based diagnosis of sickle cell anemia in human RBC." 2021 2nd International Conference on Intelligent Engineering and Management (ICIEM). IEEE, 2021.

https://doi.org/10.1109/ICIEM51511.2021.9445293