An artificial intelligence model for malaria diagnosis
Tuoyu Liu, Yanbing Li, Haidong Zhang, Ruicun Liu, Shan Yang, Yingtan Zhuang, Yue Teng · Scientia Sinica Vitae · 2022
Malaria is a mosquito-borne disease caused by Plasmodium, which has a high mortality rate. Rapid and accurate detection of malaria is the key to reducing malaria mortality and controlling its transmission. Several deep learning algorithms have previously been applied to malaria blood smears for diagnosis using features extracted from microscopic images. Here, we propose a novel artificial intelligence (AI)-based object detection model for malaria diagnosis (AIM) using multi-scale attention approaches. An annotated dataset (SmartMalariaNET) was created consisting of thin smear images acquired by smartphone cameras, which is used to train and assess the AIM. The results show that the effectiveness of our model in distinguishing positive and negative images is evident in the values of the performance metrics, namely Accuracy, Precision, Recall, F1-score, and AUC (area under curve), calculated as 94.49%, 94.54%, 94.49%, 94.50%, and 98.11%, respectively. AIM is superior to the existing deep learning models in all evaluation indicators. Our model shows clinically acceptable performance in detecting malaria parasites and could aid in malaria diagnosis in resource-limited regions, especially in areas lacking experienced parasitologists and equipment.