Deep Learning Approaches for Malaria Diagnosis: A Comparative Study of Custom CNN and Transfer Learning Models in Blood Smear Analysis

Anselme Atchogou; Abdoulmalık Abdoulmalık

EN TR

Deep Learning Approaches for Malaria Diagnosis: A Comparative Study of Custom CNN and Transfer Learning Models in Blood Smear Analysis

Abstract

The Plasmodium genus of single-celled parasites is the cause of malaria. These parasites are spread to humans through the bite of an infected Anopheles mosquito. Particularly in sub-Saharan Africa, where it severely strains health systems and economics, the illness remains an essential public health issue. Effective diagnosis and treatment are crucial for controlling and eventually eliminating malaria. The microscopic analysis of blood smears is the conventional yet labour-intensive method for diagnosing malaria, demanding extensive expertise. Automated detection through deep learning presents a vital alternative, particularly crucial for sub-Saharan Africa. This study aims to meticulously compare the performance of a custom-designed Convolutional Neural Network (CNN) with five advanced transfer learning models, ResNet50, VGG19, InceptionV3, EfficientNet-B3, EfficientNet-B7, and YOLOv11m, in categorizing segmented red blood cell images for malaria detection. Our approach involves comprehensive image preprocessing, data augmentation, and the implementation of various models. The models were evaluated using a National Library of Medicine dataset based on various metrics, including F1 Score, Accuracy, Precision, Recall, Matthews Correlation Coefficient (MCC), AUC ROC, and AUC PR. The EfficientNet-B3 model emerged as the top performer, surpassing even the custom CNN with an impressive F1 Score of 98.12%, Accuracy of 98.08%, and an MCC of 96.15%, demonstrating its superior predictive power and reliability. YOLOv11m also showed strong performance with an F1 Score of 96.89%, Accuracy of 96.93%, and MCC of 93.91%, highlighting its efficiency for real-time applications. Although the custom CNN did not outperform the advanced models, it still exhibited commendable performance, underscoring the potential of tailored architectures. The results of this study demonstrate the great potential that deep learning methods have to improve the precision of malaria diagnosis, providing notable benefits to the healthcare systems, especially for regions most severely impacted by the disease.

Keywords

Malaria diagnosis, Deep learning, CNN, Transfer learning, Healthcare

Sıtma Tanısı İçin Derin Öğrenme Yaklaşımları: Kan Sürüntüsü Analizinde Özel CNN ve Transfer Öğrenme Modellerinin Karşılaştırmalı Çalışması

Abstract

Tek hücreli parazitlerin Plasmodium cinsi, sıtmanın nedenidir. Bu parazitler, enfekte olmuş Anopheles sivrisineklerinin ısırmasıyla insanlara bulaşır. Özellikle sağlık sistemlerini ve ekonomiyi ciddi şekilde zorlayan Sahra altı Afrika'da, bu hastalık önemli bir halk sağlığı sorunu olmaya devam etmektedir. Sıtmayı kontrol altına almak ve nihayetinde ortadan kaldırmak için etkili tanı ve tedavi çok önemlidir. Kan yaymalarının mikroskobik analizi, sıtmayı teşhis etmek için geleneksel ancak emek yoğun bir yöntemdir ve kapsamlı uzmanlık gerektirir. Derin öğrenme yoluyla otomatik tespit, özellikle Sahra altı Afrika için hayati bir alternatif sunmaktadır. Bu çalışma, sıtma tespiti için segmentlere ayrılmış kırmızı kan hücresi görüntülerini sınıflandırmada, özel olarak tasarlanmış bir Convolutional Neural Network (CNN) ile beş gelişmiş transfer öğrenme modeli olan ResNet50, VGG19, InceptionV3, EfficientNet-B3, EfficientNet-B7 ve YOLOv11m'nin performansını titizlikle karşılaştırmayı amaçlamaktadır. Yaklaşımımız, kapsamlı görüntü ön işleme, veri artırma ve çeşitli modellerin uygulanmasını içermektedir. Modeller, F1 Skoru, Doğruluk, Hassasiyet, Geri Çağırma, Matthews Korelasyon Katsayısı (MCC), AUC ROC ve AUC PR gibi çeşitli metriklere dayalı olarak Ulusal Tıp Kütüphanesi veri seti kullanılarak değerlendirildi. EfficientNet-B3 modeli, %98,12'lik etkileyici bir F1 Skoru, %98,08'lik Doğruluk ve %96,15'lik MCC ile özel CNN'yi bile geride bırakarak en iyi performans gösteren model olarak öne çıktı ve üstün tahmin gücü ve güvenilirliğini kanıtladı. YOLOv11m de %96,89 F1 Skoru, %96,93 Doğruluk ve %93,91 MCC ile güçlü bir performans sergiledi ve gerçek zamanlı uygulamalar için verimliliğini vurguladı. Özel CNN, gelişmiş modelleri geride bırakmasa da, yine de övgüye değer bir performans sergiledi ve özel olarak tasarlanmış mimarilerin potansiyelini vurguladı. Bu çalışmanın sonuçları, derin öğrenme yöntemlerinin sıtma tanısının doğruluğunu artırma konusunda büyük potansiyele sahip olduğunu ve özellikle bu hastalıktan en ciddi şekilde etkilenen bölgelerde sağlık sistemlerine önemli faydalar sağladığını göstermektedir.

Keywords

Sıtma tanısı, Derin öğrenme, CNN, Transfer öğrenme, Sağlık hizmetleri

References

C. Prevention, “for DC and. CDC-Malaria-About Malaria-Biology,” CDC-Centers Dis Control Prev, 2020.
D. Posfai et al., “Plasmodium parasite exploits host aquaporin-3 during liver stage malaria infection,” PLoS Pathog, vol. 14, no. 5, p. e1007057, May 2018, doi: 10.1371/journal.ppat.1007057.10.1371/journal.ppat.1007057
E. Worrall, S. Basu, and K. Hanson, “Is malaria a disease of poverty? A review of the literature,” Tropical Medicine & International Health, vol. 10, no. 10, pp. 1047-1059, 2005.
W. H. Organization, World malaria report 2023. World Health Organization, 2023.
T. Shavlakadze et al., “Short-term Low-Dose mTORC1 Inhibition in Aged Rats Counter-Regulates Age-Re- lated Gene Changes and Blocks Age-Related Kidney Pathology,” J Gerontol A Biol Sci Med Sci, vol. 73, no. 7, pp. 845-852, Jun 14 2018, doi: 10.1093/gerona/glx249.10.1093/gerona/glx249
S. Mishra, “Malaria parasite detection using efficient neural ensembles,” Journal of Electronics, Electro- medical Engineering, and Medical Informatics, vol. 3, no. 3, pp. 119-133, 2021.
J. P. Daily, A. Minuti, and N. Khan, “Diagnosis, Treatment, and Prevention of Malaria in the US: A Review,” JAMA, vol. 328, no. 5, pp. 460-471, Aug 2 2022, doi: 10.1001/jama.2022.12366.10.1001/jama.2022.12366
S. Rajaraman et al. , “Pre-trained convolutional neural networks as feature extractors toward improved malaria parasite detection in thin blood smear images,” PeerJ, vol. 6, p. e4568, 2018.
S. Savkare and S. Narote, “Automated system for malaria parasite identification,” in 2015 international conference on communication, information & computing technology (ICCICT), 2015: IEEE, pp. 1-4.
V. K. Bairagi and K. C. Charpe, “Comparison of Texture Features Used for Classification of Life Stages of Malaria Parasite,” Int J Biomed Imaging, vol. 2016, p. 7214156, 2016, doi: 10.1155/2016/7214156.10.1155/2016/72 14156

A. Maqsood, M. S. Farid, M. H. Khan, and M. Grzegorzek, “Deep malaria parasite detection in thin blood smear microscopic images,” Applied Sciences, vol. 11, no. 5, p. 2284, 2021.
F. Sherif and A. Mohammed, “Detection of Malaria Infection Using Convolutional Neural Networks,” in 2023 Intelligent Methods, Systems, and Applications (IMSA), 2023: IEEE, pp. 291-296.
E. Irmak, “A Novel Implementation of Deep-Learning Approach on Malaria Parasite Detection from Thin Blood Cell Images,” (in English), Electrica, vol. 21, no. 2, pp. 216-224, May 2021, doi: 10.5152/electri - ca.2020.21004.10.5152/electrica.2020.21004
P. Krishnadas and N. Sampathila, “Automated detection of malaria implemented by deep learning in Py- Torch,” in 2021 IEEE International Conference on Electronics, Computing and Communication Techno - logies (CONECCT), 2021: IEEE, pp. 01-05.
M. O. F. Goni et al., “Diagnosis of Malaria Using Double Hidden Layer Extreme Learning Machine Algorit- hm With CNN Feature Extraction and Parasite Inflator,” IEEE Access, vol. 11, pp. 4117-4130, 2023.
B. Almuhaya, R. Mohammed, A. Mohammed, and B. Saha, “Malaria Detection Using Convolutional Neural Networks: A Comparative Study,” in 2023 3rd International Conference on Emerging Smart Technologies and Applications (eSmarTA), 2023: IEEE, pp. 1-8.
A. Agrawal, C. Hsu, and S. Tan, “A Comparative Analysis of Supervised and Semi-Supervised Learning Models for Malaria Classification Using Explainable AI Techniques.”
R. Nakasi, E. Mwebaze, A. Zawedde, J. Tusubira, B. Akera, and G. Maiga, “A new approach for microscopic diagnosis of malaria parasites in thick blood smears using pre-trained deep learning models,” SN Applied Sciences, vol. 2, pp. 1-7, 2020. 71 Anselme ATCHOGOU, Abdoul MALİK
Q. Quan, J. Wang, and L. Liu, “An Effective Convolutional Neural Network for Classifying Red Blood Cells in Malaria Diseases,” Interdiscip Sci, vol. 12, no. 2, pp. 217-225, Jun 2020, doi: 10.1007/s12539-020-00367- 7.10.1007/s12539-020-00367-7
V. Magotra and M. K. Rohil, “Malaria Diagnosis Using a Lightweight Deep Convolutional Neural Network,” Int J Telemed Appl, vol. 2022, p. 4176982, 2022, doi: 10.1155/2022/4176982.10.1155/2022/4176982
K. M. Pimple, P. P. Likhitkar, and S. Pande, “Convolutional neural networks for malaria image classification,” in Proceedings of Data Analytics and Management: ICDAM 2021, Volume 2, 2022: Springer, pp. 459-470.
Z. Liang et al., “CNN-based image analysis for malaria diagnosis,” in 2016 IEEE international conference on bioinformatics and biomedicine (BIBM), 2016: IEEE, pp. 493-496.
B. Lufyaila, B. Mgawe, and A. Sam, “Fine-Tuned YOLO-Based Deep Learning Model for Detecting Malaria Parasites and Leukocytes in Thick Smear Images: A Tanzanian Case Study,” Machine Learning with Appli- cations, p. 100687, 2025.
Google Colab. [Online]. Available: https:/ /colab.research.google.com
Keras. [Online]. Available: https:/ /keras.io/
TensorFlow. [Online]. Available: https:/ /www.tensorflow.org
Ultralytics YOLO11. [Online]. Available: https:/ /docs.ultralytics.com/fr/models/yolo11
Kaggle. [Online]. Available: https:/ /www.kaggle.com/
N. Jegham, C. Y. Koh, M. Abdelatti, and A. Hendawi, “Yolo evolution: A comprehensive benchmark and architectural review of yolov12, yolo11, and their previous versions,” arXiv preprint arXiv:2411.00201, 2024.
R. Khanam and M. Hussain, “Yolov11: An overview of the key architectural enhancements,” arXiv preprint arXiv:2410.17725, 2024.
G. Naidu, T. Zuva, and E. M. Sibanda, “A review of evaluation metrics in machine learning algorithms,” in Computer Science On-line Conference, 2023: Springer, pp. 15-25.
A. Pages-Zamora, M. Cabrera-Bean, and C. Diaz-Vilor, “Unsupervised online clustering and detection al - gorithms using crowdsourced data for malaria diagnosis,” Pattern Recognition, vol. 86, pp. 209-223, 2019.
D. Bibin, M. S. Nair, and P. Punitha, “Malaria parasite detection from peripheral blood smear images using deep belief networks,” IEEE Access, vol. 5, pp. 9099-9108, 2017.
S. Sarkar, R. Sharma, and K. Shah, “Malaria detection from RBC images using shallow convolutional neural networks,” arXiv preprint arXiv:2010.11521, 2020.

Details

Primary Language

English

Subjects

Artificial Intelligence (Other)

Journal Section

Research Article

Authors

Anselme Atchogou
0009-0002-1593-516X
Türkiye

Abdoulmalık Abdoulmalık
0009-0006-6525-8346
Türkiye

Publication Date

November 13, 2025

Submission Date

January 9, 2025

Acceptance Date

October 13, 2025

Published in Issue

Year 2025 Volume: 5 Number: 2

IZ

https://izlik.org/JA37WG49XR

APA

Atchogou, A., & Abdoulmalık, A. (2025). Deep Learning Approaches for Malaria Diagnosis: A Comparative Study of Custom CNN and Transfer Learning Models in Blood Smear Analysis. OMÜ Mühendislik Bilimleri Ve Teknolojisi Dergisi, 5(2), 51-71. https://izlik.org/JA37WG49XR

AMA

1.Atchogou A, Abdoulmalık A. Deep Learning Approaches for Malaria Diagnosis: A Comparative Study of Custom CNN and Transfer Learning Models in Blood Smear Analysis. OMUJEST. 2025;5(2):51-71. https://izlik.org/JA37WG49XR

Chicago

Atchogou, Anselme, and Abdoulmalık Abdoulmalık. 2025. “Deep Learning Approaches for Malaria Diagnosis: A Comparative Study of Custom CNN and Transfer Learning Models in Blood Smear Analysis”. OMÜ Mühendislik Bilimleri Ve Teknolojisi Dergisi 5 (2): 51-71. https://izlik.org/JA37WG49XR.

EndNote

Atchogou A, Abdoulmalık A (November 1, 2025) Deep Learning Approaches for Malaria Diagnosis: A Comparative Study of Custom CNN and Transfer Learning Models in Blood Smear Analysis. OMÜ Mühendislik Bilimleri ve Teknolojisi Dergisi 5 2 51–71.

IEEE

[1]A. Atchogou and A. Abdoulmalık, “Deep Learning Approaches for Malaria Diagnosis: A Comparative Study of Custom CNN and Transfer Learning Models in Blood Smear Analysis”, OMUJEST, vol. 5, no. 2, pp. 51–71, Nov. 2025, [Online]. Available: https://izlik.org/JA37WG49XR

ISNAD

Atchogou, Anselme - Abdoulmalık, Abdoulmalık. “Deep Learning Approaches for Malaria Diagnosis: A Comparative Study of Custom CNN and Transfer Learning Models in Blood Smear Analysis”. OMÜ Mühendislik Bilimleri ve Teknolojisi Dergisi 5/2 (November 1, 2025): 51-71. https://izlik.org/JA37WG49XR.

JAMA

1.Atchogou A, Abdoulmalık A. Deep Learning Approaches for Malaria Diagnosis: A Comparative Study of Custom CNN and Transfer Learning Models in Blood Smear Analysis. OMUJEST. 2025;5:51–71.

MLA

Atchogou, Anselme, and Abdoulmalık Abdoulmalık. “Deep Learning Approaches for Malaria Diagnosis: A Comparative Study of Custom CNN and Transfer Learning Models in Blood Smear Analysis”. OMÜ Mühendislik Bilimleri Ve Teknolojisi Dergisi, vol. 5, no. 2, Nov. 2025, pp. 51-71, https://izlik.org/JA37WG49XR.

Vancouver

1.Anselme Atchogou, Abdoulmalık Abdoulmalık. Deep Learning Approaches for Malaria Diagnosis: A Comparative Study of Custom CNN and Transfer Learning Models in Blood Smear Analysis. OMUJEST [Internet]. 2025 Nov. 1;5(2):51-7. Available from: https://izlik.org/JA37WG49XR

Deep Learning Approaches for Malaria Diagnosis: A Comparative Study of Custom CNN and Transfer Learning Models in Blood Smear Analysis

Abstract

Keywords

Sıtma Tanısı İçin Derin Öğrenme Yaklaşımları: Kan Sürüntüsü Analizinde Özel CNN ve Transfer Öğrenme Modellerinin Karşılaştırmalı Çalışması

Abstract

Keywords

References

Details

Primary Language

Subjects

Journal Section

Authors

Publication Date

Submission Date

Acceptance Date

Published in Issue

IZ

Cite