-
Improved Physics-informed neural networks loss function regularization with a variance-based term
Authors:
John M. Hanna,
Hugues Talbot,
Irene E. Vignon-Clementel
Abstract:
In machine learning and statistical modeling, the mean square or absolute error is commonly used as an error metric, also called a "loss function." While effective in reducing the average error, this approach may fail to address localized outliers, leading to significant inaccuracies in regions with sharp gradients or discontinuities. This issue is particularly evident in physics-informed neural n…
▽ More
In machine learning and statistical modeling, the mean square or absolute error is commonly used as an error metric, also called a "loss function." While effective in reducing the average error, this approach may fail to address localized outliers, leading to significant inaccuracies in regions with sharp gradients or discontinuities. This issue is particularly evident in physics-informed neural networks (PINNs), where such localized errors are expected and affect the overall solution. To overcome this limitation, we propose a novel loss function that combines the mean and the standard deviation of the chosen error metric. By minimizing this combined loss function, the method ensures a more uniform error distribution and reduces the impact of localized high-error regions. The proposed loss function is easy to implement and tested on problems of varying complexity: the 1D Poisson equation, the unsteady Burgers' equation, 2D linear elastic solid mechanics, and 2D steady Navier-Stokes equations. Results demonstrate improved solution quality and lower maximum error compared to the standard mean-based loss, with minimal impact on computational time.
△ Less
Submitted 26 February, 2025; v1 submitted 18 December, 2024;
originally announced December 2024.
-
A comparative analysis of metamodels for lumped cardiovascular models, and pipeline for sensitivity analysis, parameter estimation, and uncertainty quantification
Authors:
John M. Hanna,
Pavlos Varsos,
Jérôme Kowalski,
Lorenzo Sala,
Roel Meiburg,
Irene E. Vignon-Clementel
Abstract:
Zero-dimensional (0D) cardiovascular models are reduced-order models used to study global circulation dynamics and transport. They provide estimates of biomarkers (such as pressure, flow rates, and concentrations) for surgery planning and boundary conditions for high-fidelity 3D models. Although their computational cost is low, tasks like parameter estimation and uncertainty quantification require…
▽ More
Zero-dimensional (0D) cardiovascular models are reduced-order models used to study global circulation dynamics and transport. They provide estimates of biomarkers (such as pressure, flow rates, and concentrations) for surgery planning and boundary conditions for high-fidelity 3D models. Although their computational cost is low, tasks like parameter estimation and uncertainty quantification require many model evaluations, making them computationally expensive. This motivates building metamodels. In this work, we propose a pipeline from 0D models to metamodel building for tasks such as sensitivity analysis, parameter estimation, and uncertainty quantification. Three strategies are explored: Neural Networks, Polynomial Chaos Expansion, and Gaussian Processes, applied to three different 0D models. The first model predicts portal vein pressure after surgery, considering liver hemodynamics and global circulation. The second simulates whole-body circulation under pulmonary arterial hypertension before and after shunt insertion. The third assesses organ blood perfusion after revascularization surgery, focusing on contrast agent transport, requiring specific metamodel treatment. Metamodels are trained and tested on synthetic data. Neural networks proved the most efficient in terms of result quality, computational time, and ease for parameter estimation, sensitivity analysis, and uncertainty quantification. Finally, we demonstrate the full pipeline with a neural network as the emulator.
△ Less
Submitted 16 October, 2024;
originally announced October 2024.
-
Diffusion-weighted MRI-guided needle biopsies permit quantitative tumor heterogeneity assessment and cell load estimation
Authors:
Yi Yin,
Kai Breuhahn,
Hans-Ulrich Kauczor,
Oliver Sedlaczek,
Irene E. Vignon-Clementel,
Dirk Drasdo
Abstract:
Quantitative information on tumor heterogeneity and cell load could assist in designing effective and refined personalized treatment strategies. It was recently shown by us that such information can be inferred from the diffusion parameter D derived from the diffusion-weighted MRI (DWI) if a relation between D and cell density can be established. However, such relation cannot a priori be assumed t…
▽ More
Quantitative information on tumor heterogeneity and cell load could assist in designing effective and refined personalized treatment strategies. It was recently shown by us that such information can be inferred from the diffusion parameter D derived from the diffusion-weighted MRI (DWI) if a relation between D and cell density can be established. However, such relation cannot a priori be assumed to be constant for all patients and tumor types. Hence to assist in clinical decisions in palliative settings, the relation needs to be established without tumor resection. It is here demonstrated that biopsies may contain sufficient information for this purpose if the localization of biopsies is chosen as systematically elaborated in this paper. A superpixel-based method for automated optimal localization of biopsies from the DWI D-map is proposed. The performance of the DWI-guided procedure is evaluated by extensive simulations of biopsies. Needle biopsies yield sufficient histological information to establish a quantitative relationship between D-value and cell density, provided they are taken from regions with high, intermediate, and low D-value in DWI. The automated localization of the biopsy regions is demonstrated from a NSCLC patient tumor. In this case, even two or three biopsies give a reasonable estimate. Simulations of needle biopsies under different conditions indicate that the DWI-guidance highly improves the estimation results. Tumor cellularity and heterogeneity in solid tumors may be reliably investigated from DWI and a few needle biopsies that are sampled in regions of well-separated D-values, excluding adipose tissue. This procedure could provide a way of embedding in the clinical workflow assistance in cancer diagnosis and treatment based on personalized information.
△ Less
Submitted 28 February, 2021;
originally announced March 2021.
