Research on development and application of an artifical intelligence platform assisting Plasmodium detection

doi:10.12140/j.issn.1000-7423.2025.03.007

Abstract

Abstract:

Objective To develop and establish an artificial intelligence (AI) platform assisting malaria parasites detection, and the applied effect was evaluated. Methods Blood smears were collected from current cases of overseas imported malaria in Wuhan City from 2008 to 2022. Microscopic images were captured with an automatic imaging collection system, and annotated by professional microscopists to create a dataset of thin blood smear images. A detection model was employed to detect Plasmodium, white blood cells (WBCs) and red blood cells (RBCs) to deduce negativity, positivity, infection rate and infected parasite species, and the results were displayed in the application interface. Based on numbers of true positive, true negative, false positive and false negative samples, the accuracy, precision, recall, and F1 score were estimated to evaluate the performance of the model. Microscopic examination procedures were mimicked according to the technical assessment method of malaria microscopists, and 10 blood smears were randomly selected to score the model comprehensive results. Results The thin blood smears image datasets included 3 704 images that were annotated by microscopist, including 7 835 Plasmodium annotations and 786 WBC annotations. Results from the YOLOv8 confusion matrix showed an overall accuracy of 85.5%, precision of 98.5%, recall of 93.3%, F1 score of 95.8%, and an mAP@0.50 score of 96.2%. The model had a high accuracy for detection of P. vivax (96.9%) and P. malariae (93.8%), and a low accuracy for detection of P. falciparum (87.2%) and WBCs (87.8%). Except an 89.7% F1 score for detection of P. falciparum, the model had F1 scores of over 90.0% for detection of other Plasmodium species and WBCs. The recall of the model for detection of 4 Plasmodium species and WBCs, was the highest recall for detection of P. malariae (94.9%), followed by WBC (94.2%), P. vivax (94.1%), P. ovale (93.6%) and P. falciparum (92.4%). A total of 64 targets were falsely identified, with a false detection rate of 6.7%. The correction rates of the platform and professional microscopist were both 100%, and the clinical microscopy score was 90 points, with no false positives or negatives identified; however, there were false detections in identification of malaria parasite species. Conclusion The platform is effective to identify and locate malaria parasites, WBCs and RBCs, and its scoring results are approaching to the microscopist. Due to a high detection capability, the model is feasible for auxiliary diagnosis of clinical microscopy of malaria parasites.

Key words: Plasmodium detection, Object detection model, YOLOv8 model, Deep learning, Blood smear

CLC Number:

R382.31

WU Kai, XIA Qing, JIA Liming, FU Min. Research on development and application of an artifical intelligence platform assisting Plasmodium detection[J]. CHINESE JOURNAL OF PARASITOLOGY AND PARASITIC DISEASES, 2025, 43(3): 345-350.

Figures/Tables 4

Fig. 1

Fig. 2

Table 1

Table 2

References 24

[1]	Cowman AF, Healer J, Marapana D, et al. Malaria: Biology and disease[J]. Cell, 2016, 167(3): 610-624.
[2]	World Health Organization. World malaria report 2023[R]. Geneva: World Health Organization, 2023: 11-12.
[3]	国家传染病医学中心撰写组. 疟疾诊疗指南[J]. 中国寄生虫学与寄生虫病杂志, 2022, 40(4): 419-427.
	Expert Group of National Center for Infectious Diseases. Guidelines for malaria diagnosis and treatment[J]. Chin J Parasitol Parasit Dis Dec, 2022, 40(4): 419-427. (in Chinese)
[4]	夏志贵, 周水森, 汤林华. 中国消除疟疾的历程、意义、主要经验及消除后策略与展望[J]. 传染病信息, 2022, 35(1): 39-45, 59.
	Xia ZG, Zhou SS, Tang LH. History, impacts and experience of malaria elimination in China and strategies and prospects after elimination[J]. Infect Dis Inf, 2022, 35(1): 39-45, 59. (in Chinese)
[5]	Yin JH, Zhang L, Yi BY, et al. Imported malaria from land bordering countries in China: A challenge in preventing the reestablishment of malaria transmission[J]. Travel Med Infect Dis, 2023, 53: 102575.
[6]	Cao J, Newby G, Cotter C, et al. Achieving malaria elimination in China[J]. Lancet Public Health, 2021, 6(12): e871-e872.
[7]	朱国鼎, 曹俊. 朝督暮责: 定期开展评估是保持防止疟疾输入再传播能力的有效手段[J]. 中国血吸虫病防治杂志, 2023, 35(2): 113-115.
	Zhu GD, Cao J. Regular assessment is an effective approach to maintaining the capacity of prevention of re-establishment from imported malaria in China[J]. Chin J Schisto Control, 2023, 35(2): 113-115. (in Chinese)
[8]	周耀武, 张丽, 夏志贵. 疟疾跨境输入与继发传播的风险及对策[J]. 中国血吸虫病防治杂志, 2025, 37(1): 14-18.
	Zhou YW, Zhang L, Xia ZG. Risk of and response to cross-border importation and secondary transmission of malaria[J]. Chin J Schisto Control, 2025, 37(1): 14-18. (in Chinese)
[9]	张丽, 夏志贵. 2023年全国疟疾疫情特征分析[J]. 中国寄生虫学与寄生虫病杂志, 2024, 42(2): 135-139.
	Zhang L, Xia ZG. Epidemiological characteristics of malaria in China, 2023[J]. Chin J Parasitol Parasit Dis, 2024, 42(2): 135-139. (in Chinese)
[10]	张丽, 夏志贵, 李石柱. 2024年全国疟疾疫情特征分析[J]. 中国寄生虫学与寄生虫病杂志, 2025, 43(2): 162-166.
	Zhang L, Xia ZG, Li SZ. Epidemiological characteristics of malaria in China, 2024[J]. Chin J Parasitol Parasit Dis, 2025, 43(2): 162-166. (in Chinese)
[11]	Mbanefo A, Kumar N. Evaluation of malaria diagnostic methods as a key for successful control and elimination programs[J]. Trop Med Infect Dis, 2020, 5(2): 102.
[12]	王新宇, 张文宏. 中国消除疟疾流行后如何加强临床医生疟疾的诊治能力[J]. 临床内科杂志, 2022, 39(4): 217-218.
	Wang XY, Zhang WH. How to strengthen the ability of clinicians to diagnose and treat malaria after the elimination of malaria epidemic in China[J]. J Clin Intern Med, 2022, 39(4): 217-218. (in Chinese)
[13]	Maturana CR, de Oliveira AD, Nadal S, et al. Advances and challenges in automated malaria diagnosis using digital microscopy imaging with artificial intelligence tools: A review[J]. Front Microbiol, 2022, 13: 1006659.
[14]	杨涛, 杨岚兰, 杨米扬, 等. 基于改进型YOLO网络目标检测算法在乳腺肿瘤超声图像检测中的应用研究[J]. 中国医学装备, 2024, 21(9): 23-27.
	Yang T, Yang LL, Yang MY, et al. Study on the application of YOLO algorithm based on improved YOLO network in the detection of ultrasound image for breast tumor[J]. Chin Med Equip, 2024, 21(9): 23-27. (in Chinese)
[15]	Song YS, Liu GH. DCA-YOLO: Detection algorithm for YOLOv8 pulmonary nodules based on attention mechanism[J]. J Donghua Univ, 2025, 42(01): 78-87.
[16]	Varghese R, Sambath M. YOLOv8: A novel object detection algorithm with enhanced performance and robustness[C]// 2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems (ADICS). Chennai, India. IEEE, 2024: 1-6.
[17]	丰俊, 张丽, 涂宏, 等. 从消除到消除后: 中国输入性疟疾的疫情特征、挑战及防止再传播策略[J]. 中国热带医学, 2021, 21(1): 5-10.
	Feng J, Zhang L, Tu H, et al. From elimination to post-elimination: Characteristics, challenges and re-transmission preventing strategy of imported malaria in China[J]. Chin Trop Med, 2021, 21(1): 5-10. (in Chinese)
[18]	阮娇, 余树坤, 胡永峰, 等. 武汉市消除疟疾阶段境外输入性疟疾疫情与病例处置机制[J]. 公共卫生与预防医学, 2023, 34(1): 74-77.
	Ruan J, Yu SK, Hu YF, et al. Characteristics and management mechanism of imported malaria from overseas during the malaria elimination phase in Wuhan City[J]. J Public Health Prev Med, 2023, 34(1): 74-77. (in Chinese)
[19]	顾己悦, 柴丽莹, 曹园园, 等. 我国输入性疟疾患者回国后就医时机及其影响因素的系统评价[J]. 现代预防医学, 2024, 51(17): 3223-3230.
	Gu JY, Cai LY, Cao YY, et al. Systematic review of the timing of medical consultation and influencing factors for imported malaria patients returning to China[J]. Mod Prev Med, 2024, 51(17): 3223-3230. (in Chinese)
[20]	Xia J, Wu DN, Wu K, et al. Epidemiology of Plasmodium falciparum malaria and risk factors for severe disease in Hubei Province, China[J]. Am J Trop Med Hyg, 2020, 103(4): 1534-1539.
[21]	吴凯, 陈芳, 徐明星, 等. 2008—2017年武汉市境外输入性疟疾疫情及诊断能力分析[J]. 中国血吸虫病防治杂志, 2018, 30(3): 289-294.
	Wu K, Chen F, Xu MX, et al. Epidemic situation of imported malaria and diagnostic capabilities of medical institutions in Wuhan City from 2008 to 2017[J]. Chin J Schisto Control, 2018, 30(3): 289-294. (in Chinese)
[22]	Rajaraman S, Jaeger S, Antani SK. Performance evaluation of deep neural ensembles toward malaria parasite detection in thin-blood smear images[J]. Peer J, 2019, 7: e6977.
[23]	Rajaraman S, Antani SK, Poostchi M, et al. Pre-trained convolutional neural networks as feature extractors toward improved malaria parasite detection in thin blood smear images[J]. Peer J, 2018, 6: e4568.
[24]	刘拓宇, 李艳冰, 张海东, 等. 人工智能在疟原虫检测中的应用研究[J]. 中国科学: 生命科学, 2023, 53: 876-884.
	Liu TY, Li YB, Zhang HD, et al. An artificial intelligence model for malaria diagnosis[J]. Sci Sin Vitae, 2023, 53: 876-884. (in Chinese)

模型 Model	平均精度@0.50 mAP@0.50	平均精度@0.50-0.95 mAP@0.50-0.95	召回率 Recall
YOLOv8	0.962	0.762	0.933
Faster R-CNN	0.943	0.706	0.77
EfficientNet-Det	0.955	0.72	0.819

类别 Category	正确率/% Accuracy/%	精确率/% Precision/%	召回率/% Recall/%	F1分数/% F1_Score/%	平均精度@0.50mAP@0.50
恶性疟原虫 P. falciparum	88.6	87.2	92.4	89.7	93.9
间日疟原虫 P. vivax	99.0	96.9	94.1	95.4	97.3
卵形疟原虫 P. ovale	98.5	91.0	93.6	92.2	98.0
三日疟原虫 P. malariae	97.9	93.8	94.9	94.3	94.5
白细胞 White blood cell	98.6	87.8	94.2	90.8	97.2