Key result
A phenomenological model predicted platelet deposition rates with an average median error of 14.2%, outperforming mass-transfer boundary layer (21%) and machine-learning (20.7%) approaches.
Population
Ex vivo perfusion chamber experiments using blood from Large White x Landrace commercial pigs (n=4, ~36 kg).
Comparison
Three computational modeling approaches to… vs Empirical measurements of platelet deposition…
Design
Preclinical
Authors
Loading...
May enhance computational thrombosis modeling; leaves open clinical validation before device or patient use.
Absolute Event Rate: 14.2% vs 21%
A phenomenological model based on empirical facts of platelet deposition provides higher predictive accuracy than mechanistic mass-transfer or machine learning approaches.
Pallarès et al. (2015) studied Thrombosis (n=4). Phenomenological model (PM) vs. Mass-transfer boundary layer (MBL) model and Random Forest (RF) algorithm was evaluated on Median relative error in predicting platelet deposition. A phenomenological model predicted platelet deposition rates with an average median error of 14.2%, outperforming mass-transfer boundary layer (21%) and machine-learning (20.7%) approaches.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: