Abstract Background This research emphasizes the profound impact of intestinal parasitic infections on developing countries, particularly in regions such as sub-Saharan Africa, South and Central America, China, and East Asia. With over 1.5 billion people globally affected and 450 million facing serious illness and a mortality rate of 155,000 cases per year, the socio-economic hindrances posed by these infections necessitate innovative approaches for control and eradication. The integration of artificial intelligence (AI) and machine learning (ML) into parasitology, exemplified by the “Automated Diagnosis of Intestinal Parasites” (ADIP), from brazilian Portuguese “Diagnóstico Automatizado de Parasitas Intestinais” (DAPI) system, showcases a paradigm shift. This advanced system, combining simple and complex decision-making mechanisms, demonstrates promising results, achieving high agreement compared to TF-Test (three fecal test). Methods Data were extracted through reports from the laboratory information system (LIS) from January-December 2023. TF-test was replaced by ADIP system in October 2023. Data were analyzed using Microsoft Excel software, with a bibliographic survey in national and international repositories. Results The parasitology laboratory of AFIP implemented artificial intelligence (AI) with ADIP equipment associated with system LIS automation, automating the analysis of approximately 38,000 monthly stool samples. The ADIP classification, image analysis and subsequent algorithm allowed a reduced examination time, since approximately 90% of samples could be released automatically, optimizing the whole process. In addition, after a three-month period using only ADIP, positivity results increased from 7,63% (TF-test) to 9,53%. Our database is periodically reviewed by specialized professionals who validate positive results, ensuring the security of information obtained by artificial intelligence, bringing efficiency, safety, and quality to the laboratory processes. Conclusions Artificial intelligence (AI) in parasite identification and the automated results system promises a bright future. Collaboration between the scientific community and data scientists is crucial to enhance laboratory processes. Through these tools, it is possible to focus on continuous improvements, increased efficiency, and, most importantly, without sacrificing accuracy and result quality. With new technologies associated with the parasitology field, detailed records can be ensured, previously challenging to obtain, showcasing each step from result acquisition to issued report. Process improvement in parasitology brings optimizations, suggesting the potential for team reduction without compromising quality and improving the takt time.
Read full abstract