The Physics of Rubber Elasticity (J. Meixner)

Despite its great success, machine learning can have its limits when dealing with insufficient training data. A potential solution is the additional integration of prior knowledge into the training process which leads to the notion of informed machine learning. In this paper, we present a structured overview of various approaches in this field. We provide a definition and propose a concept for informed machine learning which illustrates its building blocks and distinguishes it from conventional machine learning. We introduce a taxonomy that serves as a classification framework for informed machine learning approaches. It considers the source of knowledge, its representation, and its integration into the machine learning pipeline. Based on this taxonomy, we survey related research and describe how different knowledge representations such as algebraic equations, logic rules, or simulation results can be used in learning systems. This evaluation of numerous papers on the basis of our taxonomy uncovers key methods in the field of informed machine learning.

Informed Machine Learning -- A Taxonomy and Survey of Integrating Knowledge into Learning Systems

The purpose of this chapter is to present the basic ideas of the theory of distributions. Distributions or generalized functions, as they are also known, have proved to be very useful in many branches of pure and applied mathematics. Many textbooks, monographs and articles have been written on their theory and their applications [12], [23], [53], [63], [64], [69], [71], [80], [97], [111]. Our present aim is to give a brief but solid introduction to the theory of distributions, particularly to those aspects that are important in the theory of asymptotic expansions.

Introduction to the Theory of Distributions

Manufacturing organizations need to use different kinds of techniques and tools in order to fulfill their foundation goals. In this aspect, using machine learning (ML) and data mining (DM) techniques and tools could be very helpful for dealing with challenges in manufacturing. Therefore, in this paper, a comprehensive literature review is presented to provide an overview of how machine learning techniques can be applied to realize manufacturing mechanisms with intelligent actions. Furthermore, it points to several significant research questions that are unanswered in the recent literature having the same target. Our survey aims to provide researchers with a solid understanding of the main approaches and algorithms used to improve manufacturing processes over the past two decades. It presents the previous ML studies and recent advances in manufacturing by grouping them under four main subjects: scheduling, monitoring, quality, and failure. It comprehensively discusses existing solutions in manufacturing according to various aspects, including tasks (i.e., clustering, classification, regression), algorithms (i.e., support vector machine, neural network), learning types (i.e., ensemble learning, deep learning), and performance metrics (i.e., accuracy, mean absolute error). Furthermore, the main steps of knowledge discovery in databases (KDD) process to be followed in manufacturing applications are explained in detail. In addition, some statistics about the current state are also given from different perspectives. Besides, it explains the advantages of using machine learning techniques in manufacturing, expresses the ways to overcome certain challenges, and offers some possible further research directions.

Machine learning and data mining in manufacturing

In the tissue engineering (TE) paradigm, engineering and life sciences tools are combined to develop bioartificial substitutes for organs and tissues, which can in turn be applied in regenerative medicine, pharmaceutical, diagnostic, and basic research to elucidate fundamental aspects of cell functions in vivo or to identify mechanisms involved in aging processes and disease onset and progression. The complex three-dimensional (3D) microenvironment in which cells are organized in vivo allows the interaction between different cell types and between cells and the extracellular matrix (ECM), the composition of which varies as a function of the tissue, the degree of maturation and health conditions. In this context, 3D in vitro models can more realistically reproduce a tissue or organ than two-dimensional (2D) models. Moreover, they can overcome the limitations of animal models and reduce the need for in vivo tests, according to the "3Rs" guiding principles for a more ethical research. The design of 3D engineered tissue models is currently in its development stage, showing high potential in overcoming the limitations of already available models. However, many issues are still opened, concerning the identification of the optimal scaffold-forming materials, cell source and biofabrication technology, and the best cell culture conditions (biochemical and physical cues) to finely replicate the native tissue and the surrounding environment. In the near future, 3D tissue engineered models are expected to become useful tools in the preliminary testing and screening of drugs and therapies and in the investigation of the molecular mechanisms underpinning disease onset and progression. In this review, the application of TE principles to the design of in vitro 3D models will be surveyed, with a focus on the strengths and weaknesses of this emerging approach. In addition, a brief overview on the development of in vitro models of healthy and pathological bone, heart, pancreas and liver will be presented.

/pdf/tissue-engineering-approaches-in-the-design-of-healthy-and-5fal57i4if.pdf

Tissue Engineering Approaches in the Design of Healthy and Pathological In Vitro Tissue Models.

Due to the advances in the digitalization process of the manufacturing industry and the resulting available data, there is tremendous progress and large interest in integrating machine learning and optimization methods on the shop floor in order to improve production processes. Additionally, a shortage of resources leads to increasing acceptance of new approaches, such as machine learning to save energy, time, and resources, and avoid waste. After describing possible occurring data types in the manufacturing world, this study covers the majority of relevant literature from 2008 to 2018 dealing with machine learning and optimization approaches for product quality or process improvement in the manufacturing industry. The review shows that there is hardly any correlation between the used data, the amount of data, the machine learning algorithms, the used optimizers, and the respective problem from the production. The detailed correlations between these criteria and the recent progress made in this area as well as the issues that are still unsolved are discussed in this paper.

A review of machine learning for the optimization of production processes

Modelling the mechanical behaviour of biological tissues is of vital importance for clinical applications. It is necessary for surgery simulation, tissue engineering, finite element modelling of soft tissues, etc. The theory of linear elasticity is frequently used to characterise biological tissues; however, the theory of nonlinear elasticity using hyperelastic models, describes accurately the nonlinear tissue response under large strains. The aim of this study is to provide a review of constitutive equations based on the continuum mechanics approach for modelling the rate-independent mechanical behaviour of homogeneous, isotropic and incompressible biological materials. The hyperelastic approach postulates an existence of the strain energy function--a scalar function per unit reference volume, which relates the displacement of the tissue to their corresponding stress values. The most popular form of the strain energy functions as Neo-Hookean, Mooney-Rivlin, Ogden, Yeoh, Fung-Demiray, Veronda-Westmann, Arruda-Boyce, Gent and their modifications are described and discussed considering their ability to analytically characterise the mechanical behaviour of biological tissues. The review provides a complete and detailed analysis of the strain energy functions used for modelling the rate-independent mechanical behaviour of soft biological tissues such as liver, kidney, spleen, brain, breast, etc.

https://www.degruyter.com/downloadpdf/j/bmte.2015.60.issue-6/bmt-2014-0146/bmt-2014-0146.xml

Isotropic incompressible hyperelastic models for modelling the mechanical behaviour of biological tissues: a review.

The aim of this article is to study the mechanical properties of the pancreas. Up to now, the mechanical properties of the pancreas are not sufficiently characterized. The possibility of intraoperative mechanical testing of pathological pancreata will allow the classification of pancreatic diseases in the future. The application of mechanical parameters instead of the intraoperative frozen section analysis shortens waiting times in the operating room. This study proves the general applicability of shear rheology for the determination of the mechanical properties of pancreas and the assessment of graft quality for transplantation. Porcine and human pancreas samples were examined ex vivo and a nonlinear viscoelastic behavior was observed. Pancreas was found to be more viscous than liver but both abdominal organs showed a similar flow behavior. The shear deformation dependence of healthy human pancreas was similar to porcine pancreas. An increase in the post-mortem time led to an increase in the complex modulus for a post-mortem time up to 8.5 days. Histological investigations showed that an increased amount of collagen coincides with the stiffening of the pancreatic tissue.

Experimental analysis of the mechanical behavior of the viscoelastic porcine pancreas and preliminary case study on the human pancreas.

Purpose: For a better understanding of the effects of thermally altered soft tissue, the biothermomechanics of these tissues need to be studied. Without the knowledge of the underlying physical processes and the parameters that can be controlled clinically, thermal treatment of cancerous hepatic tissue or the preservation of liver grafts are based primarily on trial and error. Materials and methods: Thus, this study is concerned with the investigation of the influence of temperature on the rheological properties and the histological properties of porcine liver. Results: Heating previously cooled porcine liver tissue above 40 °C leads to significant, irreversible stiffness changes observed in the amplitude sweep. The increase of the complex shear module of healthy porcine liver from room temperature to 70 °C is approximately 9-fold. Comparing the temperatures −20 °C and 20 °C, no significant difference of the mechanical properties was observed. Furthermore, there is a strong relation between the me...

https://www.tandfonline.com/doi/pdf/10.3109/02656736.2014.924161

Anke Stoll

Papers

A review of machine learning for the optimization of production processes

Isotropic incompressible hyperelastic models for modelling the mechanical behaviour of biological tissues: a review.

Experimental analysis of the mechanical behavior of the viscoelastic porcine pancreas and preliminary case study on the human pancreas.

Mechanics of fresh, frozen-thawed and heated porcine liver tissue.

How preservation time changes the linear viscoelastic properties of porcine liver.