Key result
An intelligent platelet aggregate classifier (iPAC) based on a convolutional neural network successfully classified platelet aggregates by agonist type with an average accuracy of 77%.
Why the study?
Platelet aggregation is elicited by various agonists, but these aggregates have long been considered indistinguishable and impossible to classify.
Population
Blood samples from 5 healthy human subjects (1 for initial training/testing, 4 for validation)
Comparison
Intelligent platelet aggregate classifier using… vs Conventional flow cytometer
Design
Preclinical
Authors
Loading...
May aid preclinical thrombosis studies; leaves open human diagnostic translation.
A novel deep learning-based imaging flow cytometry method (iPAC) can successfully classify platelet aggregates by their activating agonist, offering a new tool for studying thrombosis and potential diagnostic applications.
Zhou et al. (2020) studied Healthy subjects (platelet aggregation analysis) (n=5). Intelligent platelet aggregate classifier (iPAC) vs. Conventional flow cytometry was evaluated on Classification accuracy of platelet aggregates by agonist type. An intelligent platelet aggregate classifier (iPAC) based on a convolutional neural network successfully classified platelet aggregates by agonist type with an average accuracy of 77%.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: