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

• Published in: IEEE open journal of engineering in medicine and biology Vol. 4; pp. 226 - 233
• Main Authors: Zhong, Yuming, Dan, Ying, Cai, Yin, Lin, Jiamin, Huang, Xiaoyao, Mahmoud, Omnia, Hald, Eric S., Kumar, Akshay, Fang, Qiang, Mahmoud, Seedahmed S.
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.01.2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Africa; Algorithms; Artificial neural networks; Blood; Computer-aided diagnosis; Datasets; Graphical user interface; Image acquisition; Image processing; Image quality; Image segmentation; Leukocytes; Malaria; malaria parasites; Medical diagnostic imaging; Medical imaging; microscope; Microscopes; Microscopy; Model accuracy; Neural networks; Parasites; Smart phones; Smartphones; Vector-borne diseases; Sub-Saharan Africa; smartphones; image segmentation; Computer-aided diagnosis; malaria parasites; microscope; neural networks
• ISSN: 2644-1276; 2644-1276
• DOI: 10.1109/OJEMB.2023.3328435

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.