Ulas Cilingir

Efficient probabilistic model personalization integrating uncertainty on data and parameters: Application to Eikonal-Diffusion models in cardiac electrophysiology

Biophysical models are increasingly used for medical applications at the organ scale. However, model predictions are rarely associated with a confidence measure although there are important sources of uncertainty in computational physiology methods. For instance, the sparsity and noise of the clinical data used to adjust the model parameters (personalization), and the difficulty in modeling accurately soft tissue physiology. The recent theoretical progresses in stochastic models make their use computationally tractable, but there is still a challenge in estimating patient-specific parameters with such models. In this work we propose an efficient Bayesian inference method for model personalization using polynomial chaos and compressed sensing. This method makes Bayesian inference feasible in real 3D modeling problems. We demonstrate our method on cardiac electrophysiology. We first present validation results on synthetic data, then we apply the proposed method to clinical data. We demonstrate how this can help in quantifying the impact of the data characteristics on the personalization (and thus prediction) results. Described method can be beneficial for the clinical use of personalized models as it explicitly takes into account the uncertainties on the data and the model parameters while still enabling simulations that can be used to optimize treatment. Such uncertainty handling can be pivotal for the proper use of modeling as a clinical tool, because there is a crucial requirement to know the confidence one can have in personalized models.

Effect of depth on seismic response of circular tunnels

Tunnels in seismically active areas are vulnerable to adverse effects of earthquake loading. Recent seismic events have shown that there is a need to validate current design methods to better understand the deformation mechanisms associated with the dynamic behaviour of tunnels. The research described in this paper consists of physical and numerical modelling of circular tunnels with dynamic centrifuge experiments and complementary finite element simulations. The aim is to develop an understanding of the effects of tunnel depth on the seismic behaviour of tunnels. Tunnels with different depth-to-diameter ratios were tested in dry, loose silica sand. Accelerations around the tunnel and earth pressures on the lining were measured. A high-speed digital camera was used to record soil and lining deformations. Particle image velocimetry analyses were carried out on the recorded images to measure the deformations. Complementary dynamic finite element simulations were also conducted with a code capable of managin...

Seismic behaviour of anchored quay walls with dry backfill

Abstract The seismic behaviour of anchored sheet pile walls is a complex soil-structure interaction problem. Damaged sheet pile walls are very expensive to repair and their seismic behaviour needs to be investigated in order to understand their possible mechanisms of failure. The research described in this paper involves both centrifuge testing and Finite Element (FE) analyses aimed at investigating the seismic behaviour of an anchored sheet pile wall in dry sand. The model wall is tied to the backfill with two tie rods connected to an anchor beam. The accelerations of the sheet pile wall, the anchor beam and the soil around the wall were measured using miniature piezoelectric accelerometers. The displacement at the tip of the wall was also measured. Stain gauges at five different locations on the wall were used to measure the bending moments induced in the the wall. The anchor forces in the tie rods were also measured using load cells. The results from the centrifuge tests were compared with 2-D, plane strain FE analyses conducted using DIANA-SWANDYNE II and the observed seismic behaviour was explained in the light of these findings.

Cross-Facility Validation of Dynamic Centrifuge Testing

This paper compares the results of dynamic centrifuge tests on shallow foundations conducted at two different geotechnical facilities, IFSTTAR (Institut francais des sciences et technologies des transports, de l’amenagement et des reseaux, formerly LCPC), France and Cambridge University, U.K. Both facilities ran tests on a single degree of freedom model structure with its shallow foundation located on dry sand and subjected to dynamic shaking. Measurements were taken at both facilities allowing direct comparisons to be made. Fundamentally the results obtained were found to agree well via comparison of soil amplification profiles and moment-rotation cycles. However, higher frequency components agreed less favourably. This variation is thought to be due to a mismatch between the dynamic properties of the model containers. In this series of tests the higher frequency components are not of great importance and therefore the variation is insignificant, however, in future tests when earthquake signals with higher frequency components are input it may become an issue requiring further investigation.

FLIQ: Experimental Verification of Shallow Foundation Performance Under Earthquake-Induced Liquefaction

The seismic performance of a square footing, resting on a liquefiable sand layer, with a non-liquefiable clay crust, is examined herein, with the aid of three centrifuge experiments. Emphasis is given on the seismic settlements of the foundation, while it is for the first time attempted to measure its (degraded) post-shaking bearing capacity, with the aid of a hydraulic piston, specially programmed to push the footing to failure, immediately after the end of shaking and before the dissipation of excess pore pressures. Aimed to examine the effect of clay crust thickness H on foundation performance, the experiments were performed for H = 2/3B, B and 5/3B, with B being the width of the footing. Following a brief presentation of the testing configuration, soil properties and excitation characteristics, the experimental results are presented and evaluated through comparison with relevant numerical and analytical predictions. Thus, the beneficial effect of the clay crust thickness H is quantitatively substantiated and the existence of a “critical clay crust thickness”, beyond which subsoil liquefaction does not deter foundation performance, is experimentally verified.