Malaria is a disease caused by blood infection with Plasmodium parasites imposing a heavy toll on global public health. The gold standard diagnostic method is the microscopy analysis of a blood smear, but the method is difficult to implement in low-resource settings due to the lack of equipment and skilled personnel. In this paper, an open-source image acquisition tool is presented in combination with a pipeline for image analysis of thin blood smears to assess sample quality, a critical aspect of the preparatory procedure which is often overlooked. Quantitative parameters on red blood cell morphology and distribution were extracted and correlated with the quality of the thin blood smear. Images acquired from healthy donor samples were used to test the pipeline. The pipeline proved capable of correctly extracting sample quality parameters and will be used in future developments as the basis for an artificial intelligence algorithm for an automatic classification between samples of low and high quality in diagnosis process.
Open-source toolkit for image acquisition and quality assessment of thin blood smears for malaria diagnosis
Coro, Florinda;Mangano, Valentina;Ahluwalia, Arti;De Maria, Carmelo
Ultimo
2025-01-01
Abstract
Malaria is a disease caused by blood infection with Plasmodium parasites imposing a heavy toll on global public health. The gold standard diagnostic method is the microscopy analysis of a blood smear, but the method is difficult to implement in low-resource settings due to the lack of equipment and skilled personnel. In this paper, an open-source image acquisition tool is presented in combination with a pipeline for image analysis of thin blood smears to assess sample quality, a critical aspect of the preparatory procedure which is often overlooked. Quantitative parameters on red blood cell morphology and distribution were extracted and correlated with the quality of the thin blood smear. Images acquired from healthy donor samples were used to test the pipeline. The pipeline proved capable of correctly extracting sample quality parameters and will be used in future developments as the basis for an artificial intelligence algorithm for an automatic classification between samples of low and high quality in diagnosis process.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.