Adaptive Neuro-Fuzzy Inference System (ANFIS) predicted Shore A hardness with R2 0.987, RMSE 0.701, and MAE 0.250, outperforming linear regression (R2 0.732, RMSE 7.266, MAE 5.499) in an in vitro vascular model simulating arterial stiffness.
Can machine learning approaches accurately estimate pulse wave velocity and stiffness parameters in an in vitro vascular model?
Machine learning models can accurately estimate pulse wave velocity and vascular stiffness in a controlled in vitro setting, serving as a validation step toward non-invasive in vivo applications.
Effect estimate: R2 0.987 for ANFIS vs 0.732 for linear regression
Absolute Event Rate: 0.987% vs 0.732%
p-value: p=<0.001
This study evaluates the feasibility of estimating stiffness-related parameters and pulse wave velocity (PWV) in a controlled in vitro circulatory setup using artificial silicone vessels with systematically varied Shore A hardness and wall thickness. From synchronized pressure and capacitive waveforms, fiducial points and engineered features are extracted, together with pump settings (stroke volume and heart rate). A Sugeno-type adaptive neuro-fuzzy inference system (ANFIS) is used for hardness-level prediction and benchmarked against linear regression and contemporary machine-learning/deep-learning baselines using stratified cross-validation. PWV estimates derived via hardness-to-elasticity conversion models and the Moens–Korteweg formulation are evaluated against a reference PWV obtained within the same experimental configuration. Under these controlled conditions, the proposed pipeline shows strong agreement with reference labels and measurements. The results should be interpreted as an in vitro validation step; translation to biological tissues or in vivo data will require external validation, calibration of material-property mapping, and robustness testing under physiological variability and measurement noise.
Barvík et al. (Fri,) conducted a other in Artificial arterial segments with varied wall thickness and Shore A hardness in an in vitro vascular model simulating physiological and pathological cardiovascular conditions (n=90). Adaptive Neuro-Fuzzy Inference System (ANFIS) modeling for Shore A hardness prediction and PWV estimation vs. Linear regression and contemporary ML/DL baseline models was evaluated on Prediction of Shore A hardness levels and pulse wave velocity (PWV) estimation accuracy (R2 0.987 for ANFIS vs 0.732 for linear regression, p=<0.001). Adaptive Neuro-Fuzzy Inference System (ANFIS) predicted Shore A hardness with R2 0.987, RMSE 0.701, and MAE 0.250, outperforming linear regression (R2 0.732, RMSE 7.266, MAE 5.499) in an in vitro vascular model simulating arterial stiffness.