|
|
Background: Malaria remains a significant public health challenge globally. Traditionally, microscopic examination of peripheral blood smears has been the gold standard for diagnosing malaria. However, this method not only requires a high level of professional skills from laboratory personnel but also has detection results that are sus-ceptible to various factors. In recent years, artificial intelligence (AI) technology has opened up new possibilities for the automated identification of Plasmodium parasites.
Methods: This study reporting a case of imported malaria, demonstrates the practical application of the DeepSeek model in analyzing Plasmodium parasites in peripheral blood smears and compares its results with those of traditional peripheral blood smear microscopy and Plasmodium genetic testing.
Results: In terms of accuracy, DeepSeek successfully identified Plasmodium parasites in the images of the patient's peripheral blood smears, including the morphology of their different developmental stages, thus effectively reducing the risks of misdiagnosis and missed diagnosis caused by subjective judgment and poor sample preparation quality in traditional methods. Moreover, in terms of efficiency, DeepSeek only took a few minutes to complete the detection of smear images, while traditional microscopy required experienced doctors to spend several hours or even longer.
Conclusions: The high precision and high efficiency demonstrated by DeepSeek not only fully prove its significant advantages in the field of malaria diagnosis but also provide solid support for timely treatment and disease control of patients. Moreover, its working mode of 'AI primary screening + manual review' is expected to become an important technical support for grassroots medical institutions to prevent and control imported malaria.
DOI: 10.7754/Clin.Lab.2025.250408
|
