Using Artificial Intelligence and Machine Learning to Analyze and Recognize Microscopic Images of Pathogenic Protozoa
Dr. Mohammad Taghi Ahady PhD of Medical Parasitology
ABSTRACT
Background and Aims: Accurate diagnosis of pathogenic protozoan using microscopic images is one of the important challenges in medical parasitology. Analyzing the microscopic images of infectious protozoan using the Artificial Intelligence technology are going to be the precise and powerful instrument for diagnosing of the pathogenic protozoan morphology.
Method: In this study, all the research articles published among 2015-2025 were collected by referring the valuable data bases including PubMed, Scopus, and IEEE Xplore. Criteria for entering the study was using AI and ML algorithms for analyzing of pathogenic protozoan microscopic images (PPMI). Totally, 45 related articles were studied.
Results: The results showed that Deep Learning algorithms including Convolutional Neural Networks (CNNs) have demonstrated high accuracy in diagnosis and analysis of PPMI. The accuracy of these algorithms including CNNs-based models for analyzing and distinguishing the microscopic images of Plasmodium species (malaria disease causing agents) in blood smears was over than 97%. Using image preprocessing and data augmentation techniques have led to improved performance of DL models in diagnosis and analyzing the microscopic images of Leishmania, Trypanosoma, Toxoplasma, Babesia, and Trichomonas species with over than 98% accuracy. On the other hand, AI models have been successfully used in the classification of morphological stages (trophozoites, cysts, and oocysts) of Giardia, and Cryptosporidium species by analyzing the microscopic images.
Conclusion: The technologies of AI, ML, and DL have provided great opportunity to improve the diagnostic methods of pathogenic protozoan parasites through the analysis of microscopic images. Keywords: Artificial Intelligence (AI), Machine Learning (ML), Deep Learning (DL), Pathogenic Protozoan, Diagnosis and Analyzing the Microscopic Images.
