Large language models (LLMs) have performed well across various clinical natural language processing tasks, despite not being directly trained on electronic health record (EHR) data. In this work, we examine how popular open-source LLMs learn clinical information from large mined corpora through two crucial but understudied lenses: (1) their interpretation of clinical jargon, a foundational…
OBJECTIVES: To develop the first core Critical Care Data Dictionary (C2D2) with common data elements (CDEs) to characterize critical illness and injuries. DESIGN: Group consensus process using modified Delphi approach. SETTING: Electronic surveys and in-person meetings. SUBJECTS: A multidisciplinary workgroup of clinicians and researchers with expertise in the care of the critically ill and injured….
Diffuse correlation spectroscopy (DCS) allows label-free, non-invasive investigation of microvascular dynamics deep within tissue, such as cerebral blood flow (CBF). However, the signal-to-noise ratio (SNR) in DCS limits its effective cerebral sensitivity in adults, in which the depth to the brain, through the scalp and skull, is substantially larger than in infants.Therefore, we aim to…
Monitoring disease development in arteries, which supply oxygen and nutrients to the body, is crucial and can be assessed using hemodynamic metrics. Hemodynamic metrics can be calculated via computational fluid dynamic simulation of patient-specific geometries. These simulations are known to be heavily influenced by boundary conditions, such as time-dependent inlet flow. However, the effects of…
In this Note, we address the construction of a class of stochastic Ogden’s stored energy functions associated with incompressible hyperelastic materials. The methodology relies on the maximum entropy principle, which is formulated under constraints arising in part from existence theorems in nonlinear elasticity. More specifically, constraints related to both polyconvexity and consistency with linearized elasticity…
This paper is devoted to the modeling of compressible hyperelastic materials whose response functions exhibit uncertainties at some scale of interest. The construction of parametric probabilistic representations for the Ogden class of stored energy functions is specifically considered and formulated within the framework of Information Theory. The overall methodology relies on the principle of maximum…
In this work, we address the constitutive modeling, in a probabilistic framework, of the hyperelastic response of soft biological tissues. The aim is on the one hand to mimic the mean behavior and variability that are typically encountered in the experimental characterization of such materials, and on the other hand to derive mathematical models that…
This paper deals with the construction of random field models for spatially-dependent anisotropic strain energy functions indexed by complex geometries. The approach relies on information theory and the principle of maximum entropy, which are invoked in order to construct the family of first-order marginal probability distributions in accordance with fundamental constraints such as polyconvexity, coerciveness…
This paper develops a stochastic model for the spatially-dependent material parameters parameterizing anisotropic strain energy density functions. The construction is cast within the framework of information theory, which is invoked to derive a least-informative model while ensuring consistency with theoretical requirements in finite elasticity. Specifically, almost sure polyconvexity and uniform growth conditions are enforced through…
