The Correlation of AI Based Automated Hematology Analyzers and Manual Results of Complete Blood Count in Dog and Cat Blood

The complete blood count (CBC) is a diagnostic test to analyze abnormalities of blood cells. Currently, automated hematology analyzers and artificial intelligence technology are being used with automated blood analyzers to ensure accuracy and reliability. This study aimed to evaluate the performance of artificial intelligence (AI) based automated blood cell analyzer, Awalife AI-100Vet Multifunctional Morphological Analyzer, in dog and cat blood samples by comparison with the CBC manual method. In dogs, PCV, hemoglobin, RBC, MCH, WBC, % Neutrophil, %Lymphocyte, %Monocyte, %Eosinophil and %Reticulocyte were all significantly correlated. While in cats, PCV, Hemoglobin RBC, WBC, % Neutrophil, % Lymphocyte, and % Eosinophil were all also significantly correlated. AUC values obtained by the Awalife AI-100Vet analyzer for Hematology testing in dogs and cats were 0.72 and 0.92 respectively. These findings suggest that the Awalife AI-100Vet analyzer demonstrated good accuracy using dog blood for hematology testing as well as excellent accuracy when using cat blood. The AI-based automated blood analyzer has the potential to analyze hematological data and is close to the reference method. However, there are still differences in some parameters. Further optimization of the AI algorithm, which will involve increasing the accuracy of identifying unusual cell shapes, improving stability against various samples, such as stains, and achieving good results when working with unique pathologies, should be carried out.