Frontiers in microbiology · 2025
Rapid and accurate recognition of erythrocytic stage parasites of Plasmodium falciparum via a deep learning-based YOLOv3 platform.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
He Wei, Zhu Huiyin, Geng Junjie, Zhu Daiqian, Wu Kai, Xie Li, Li Jian, Yang Hailin
Аннотация
BACKGROUND: Malaria remains a fatal global infectious disease, with the erythrocytic stage of Plasmodium falciparum being its main pathogenic phase. Early diagnosis is critical for effective treatment. This study developed and evaluated an artificial intelligence-assisted diagnosis (AI-assisted diagnostic) tool for malaria parasites.
METHODS: The peripheral blood samples of malaria patients were collected. Thin blood film smear were prepared, stained and examined by microscopic. After manual confirmation and validation with qPCR, the images of infected red blood cells (iRBCs) of P. falciparum were captured. Using a sliding window method, each original image was cropped into 20 small images (518 × 486 pixels). Selected iRBCs were classified, and P. falciparum was detected using the YOLOv3 deep learning-based object detection algorithm.
RESULTS: A total of 262 images were tested. The YOLOv3 model detected 358 P. falciparum-containing iRBCs, with a false negative rate of 1.68% (6 missed iRBCs) and false positive rate of 3.91% (14 misreported iRBCs), yielding an overall recognition accuracy of 94.41%.
CONCLUSION: The developed AI-assisted diagnostic tool exhibits robust efficiency and accuracy in Plasmodium falciparum recognition in clinical thin blood smears. It provides a feasible technical support for malaria control in resource-limited settings.
METHODS: The peripheral blood samples of malaria patients were collected. Thin blood film smear were prepared, stained and examined by microscopic. After manual confirmation and validation with qPCR, the images of infected red blood cells (iRBCs) of P. falciparum were captured. Using a sliding window method, each original image was cropped into 20 small images (518 × 486 pixels). Selected iRBCs were classified, and P. falciparum was detected using the YOLOv3 deep learning-based object detection algorithm.
RESULTS: A total of 262 images were tested. The YOLOv3 model detected 358 P. falciparum-containing iRBCs, with a false negative rate of 1.68% (6 missed iRBCs) and false positive rate of 3.91% (14 misreported iRBCs), yielding an overall recognition accuracy of 94.41%.
CONCLUSION: The developed AI-assisted diagnostic tool exhibits robust efficiency and accuracy in Plasmodium falciparum recognition in clinical thin blood smears. It provides a feasible technical support for malaria control in resource-limited settings.