Efficient Malaria Parasite Detection From Diverse Images of Thick Blood Smears for Cross-Regional Model Accuracy

Yuming Zhong; Ying Dan; Yin Cai; Jiamin Lin; Xiaoyao Huang; Omnia Mahmoud; Eric S Hald; Akshay Kumar; Qiang Fang; Seedahmed S Mahmoud

doi:10.1109/OJEMB.2023.3328435

Efficient Malaria Parasite Detection From Diverse Images of Thick Blood Smears for Cross-Regional Model Accuracy

IEEE Open J Eng Med Biol. 2023 Oct 30:4:226-233. doi: 10.1109/OJEMB.2023.3328435. eCollection 2023.

Authors

Yuming Zhong^{1

2}, Ying Dan^{1

2}, Yin Cai^{1

2}, Jiamin Lin^{1

2}, Xiaoyao Huang³, Omnia Mahmoud⁴, Eric S Hald^{1

2}, Akshay Kumar^{1

2}, Qiang Fang^{1

2}, Seedahmed S Mahmoud^{1

2}

Affiliations

¹ Department of Biomedical Engineering, College of EngineeringShantou University Shantou 515063 China.
² The Frontier Technology Research InstituteFirst Affiliated Hospital of Shantou University Shantou 515063 China.
³ Shantou University Medical CollegeShantou University Shantou 515063 China.
⁴ Alkawa Hospital Alkawa 28815 Sudan.

Abstract

Goal: The purpose of this work is to improve malaria diagnosis efficiency by integrating smartphones with microscopes. This integration involves image acquisition and algorithmic detection of malaria parasites in various thick blood smear (TBS) datasets sourced from different global regions, including low-quality images from Sub-Saharan Africa. Methods: This approach combines image segmentation and a convolutional neural network (CNN) to distinguish between white blood cells, artifacts, and malaria parasites. A portable system integrates a microscope with a graphical user interface to facilitate rapid malaria detection from smartphone images. We trained the CNN model using open-source data from the Chittagong Medical College Hospital, Bangladesh. Results: The validation process, using microscopic TBS from both the training dataset and an additional dataset from Sub-Saharan Africa, demonstrated that the proposed model achieved an accuracy of 97.74% ± 0.05% and an F1-score of 97.75% ± 0.04%. Remarkably, our proposed model with AlexNet surpasses the reported literature performance of 96.32%. Conclusions: This algorithm shows promise in aiding malaria-stricken regions, especially those with limited resources.

Keywords: Computer-aided diagnosis; image segmentation; malaria parasites; microscope; neural networks; smartphones.

Grants and funding

This work was supported by the Li Ka Shing Foundation Cross-Disciplinary Research under Grant 2020LKSFG04C.