Intelligent Classification of White Blood Cells

AI-driven Differential Blood Cell Count

AI-driven image acquisition and processing of various types of white blood cells.
The software display of Metafer allows for easy reviewing and cell counting.
Blood specimen stained with Wright’s stain are supported.
Automated, fast, and standardized workflow.
High-throughput scanning is configurable for up to 800 slides.
Circa 3 minutes scanning time per slide with suitable specimen and hardware.

White blood cells (WBCs) differential count is a widely performed hematological test. MetaSystems has developed a new algorithm based on a Deep Neural Network (DNN) for AI-assisted classification of WBCs.

Artificial Intelligence for Differential Blood Cell Count

White blood cells (WBCs) consist of various different cell types. In a blood differential test, the amount of cells per cell type are counted in a blood smear. By visually inspecting cells, even subtle morphological changes can be evaluated.

Due to morphological variations in white blood cells, visual inspection requires well-trained laboratory personnel and is usually time-consuming. Innovations in deep learning, a subset of artificial intelligence (AI), uses images of white blood cells classified by an expert to train Deep Neural Networks (DNNs).

Fully Automated Imaging Workflow

With MetaSystems, slide scanning, image acquisition, and image processing can be executed as a fully automated workflow. Sophisticated imaging hardware and software algorithms based on Deep Neural Networks (DNNs) provide a standardized procedure, high image quality, and fast image processing.

Fast Scanning and Results Review

The automated scanning workflow can achieve a speed of around 3 minutes per slide, when performed with suitable specimen and imaging hardware. The review of cell classes can be supported with a convenient, external keyboard by pressing single keystrokes (RapidScore).

Would you like to learn more about our WBC solutions?

Contact Us