Interactive Animation and Modeling by Drawing: Pedagogical Applications in Medicine

Thesis (PDF Available) · January 2003with 114 Reads
Thesis for: PhD, Advisor: Marie-Paule Cani
Abstract
Medicine is a discipline where visualization is an essential component of learning. However, the three-dimensional, dynamic structure of the human body poses difficult teaching challenges. There is a need for truly interactive computer tools that will enable students to create and manipulate computer models, not just watch them. We propose different approaches with that goal in mind. We were first interested in interactive physically-based animation of anisotropic elastic materials. One possible application scenario is an anatomy course on heart physiology where students can build interactive samples of cardiac muscular tissue. To achieve this, our model exhibits two key features. The first one is a low computational cost that results in high frame rates; the second one is an intuitive \emph(system image) that ensures easy control by the user. Next, we were interested in interaction in three dimensions using two-dimensional input, either for annotating existing models, or for creating new models; taking advantage of the fact that drawing practice is still considered a fundamental learning method by some anatomy teachers in the French medical school curriculum. Our 3D drawing system has a stroke representation that enables drawing redisplay when the viewpoint changes. Moreover, this representation can be mixed freely with existing polygonal surfaces for annotation purposes. In contrast, our modeling by drawing tool uses information from both stroke geometry and the drawn image, to allow three-dimensional modeling without explicit depth specification.

Supplementary resources

  • ... Mass-spring systems have been used to model tex- tiles [11], [14], [23], long animals such as snakes, or soft organic tissues, like muscles, face or abdomen, where the cutting of tissue can be simulated [16], [17], [19], [20] . Moreover, these systems have been used to describe a wide range of different elastic behaviors such as anisotropy [3], heterogeneity [24], non linearity [4] and also incompressibility [22]. However, where FEMs are built upon elastic theory, mass-spring models are generally far from being accurate . ...
    ... Despite of its interest, this approach requires pre-computations and the storage of an extensive amount of information for each mesh component (vertex, edge, face, element). The second approach has been proposed by Van Gelder [25] and has been referenced in [3], [5], [7], [15], [20], [26]. In this approach, Van Gelder proposes a new formulation for triangular meshes, allowing the calculation of springs stiffness constant according to elastic parameters of the object to simulate (Young's modulus E, and Poisson's ratio ν). ...
    ... Mass-spring systems have been used to model textiles [11], [14], [23], long animals such as snakes, or soft organic tissues, like muscles, face or abdomen, where the cutting of tissue can be simulated [16], [17], [19], [20]. Moreover, these systems have been used to describe a wide range of different elastic behaviors such as anisotropy [3], heterogeneity [24], non linearity [4] and also incompressibility [22]. ...
    Article
    Full-text available
    Besides finite element method, mass-spring system is widely used in Computer Graphics. It is indubitably the simplest and most intuitive deformable model that takes into account elastic considerations. This discrete model allows to perform with ease interactive deformations as well as to handle complex interactions. Thus, it is perfectly adapted to generate visually plausible animations. However, a drawback of this simple formulation is the relative difficulty to control efficiently realistic physically-based behaviors. Indeed, none of the existing models has succeeded in dealing with this satisfyingly. Moreover, we demonstrate that the mostly cited technique in the literature, proposed by Van Gelder, is far to be exact in most real cases, and consequently, this model can not be used in simulation. So, we propose a new general 3D formulation that reconstructs the geometrical model as an assembly of elementary hexahedral "bricks". Each brick (or element) is then transformed into a mass-spring system. Edges are replaced by springs that connect masses representing the vertices. The key point of our approach is the determination of the stiffness springs to reproduce the correct mechanical properties (Young's modulus and Poisson's ratio) of the reconstructed object. We validate our methodology by performing some numerical experiments. Finally, we evaluate the accuracy of our approach, by comparing our results with the deformation obtained by finite element method. Index Terms—Discrete Modeling, Physical Simulation, Mass-Spring System, Rheological Parameters. F
  • ... Mass-spring system have been used to model tex- tiles [12,14, 27], long animals such as snakes [18], or soft organic tissues, such as muscles, face or abdomen, with sometimes the possibility to simulate tissue cut [1, 7,16, 17,20,21, 26]. Moreover, these systems have been used to describe a wide range of different elastic behaviors such as anisotropy [4], heterogeneity [28], non linearity [5] and also incompressibility [24,25]. An important problem of these models is to choose an appropriate meshing that describes well the object and that does not contain any privileged direction for the strains propagation. ...
    ... But, despite its interest, this approach requires pre-computations and the storage of an extensive amount of information for each mesh component (vertex, edge, face, element). The second approach proposed by Van Gelder [29], referenced in [4, 6,9,15,21,22,30], introduces mechanical parameters (Young's modulus and Poisson's ratio) in a simple mass-spring system. This approach combines the advantages of an accurate mechanical parameterization with a hyper-elastic model, enabling either small or large deformations. ...
    Article
    Full-text available
    Besides the finite element method, the mass-spring discrete modeling is widely used in computer gra-phics. This discrete model allows to perform very eas-ily interactive deformations and to handle quite complex interactions with only a few equations. Thus, it is per-fectly adapted to generate visually correct animations. However, a drawback of this simple formulation is the relative difficulty to control efficiently physically realis-tic behaviors. Indeed, none of the existing models has succeeded to deal satisfyingly with this. Moreover, we demonstrate that the mostly cited technique in the liter-ature, proposed by Van Gelder, is far to be exact in most real cases and its interest is limited to some specific non realistic animations. Here, we propose a new general 2D formulation that reconstructs the geometrical model as an assembly of elementary "bricks". Each brick (or el-ement) is then transformed into a mass-spring system, in which edges are springs connecting masses placed on the element vertices. The key point of our approach is the determination of the stiffness constant of each spring to reproduce the correct mechanical properties (Young's modulus, Poisson's ratio and shear modulus) of the re-constructed object. We validate our methodology with the help of some numerical experimentation of mechan-ics, like stretching, shearing and loading and then we evaluate the accuracy limits of our approach.
  • Article
    Full-text available
    Besides finite element method, mass-spring discrete modeling is widely used in computer graphics. This discrete model allows to perform very easily interactive deformations and to handle quite complex interactions with only a few equations. Thus, it is perfectly adapted to generate visually correct animations. However, a drawback of this simple formulation is the relative difficulty to control efficiently physically realistic behaviors. Indeed, none of the existing models has succeeded in dealing with this satisfyingly. Here, we propose a new general 3D formulation that reconstructs the geometrical model as an assembly of elementary "bricks". Each brick (or element) is then transformed into a mass-spring system. Edges are replaced by springs that connect masses representing the ele-ment vertices. The key point of our approach is the determination of the spring-stiffness to reproduce the correct mechanical properties (Young's modulus, Poisson's ratio, bulk and shear modulus) of the reconstructed object. We validate our methodology by performing some numerical experimentations, like shearing and loading, or beam deflection and then we evaluate the accuracy limits of our approach.
  • Article
    Full-text available
    Besides finite element method, mass-spring systems are widely used in Computer Graphics. It is indubitably the simplest and most intuitive deformable model. This discrete model allows to perform interactive deformations with ease and to handle complex interactions. Thus, it is perfectly adapted to generate visually plausible animations. However, a drawback of this simple formulation is the relative difficulty to control efficiently physically realistic behaviours. Indeed, none of the existing models has succeeded in dealing with this satisfyingly. We demonstrate that this restriction cannot be overpassed with the classical mass-spring model, and we propose a new general 3D formulation that reconstructs the geometrical model as an assembly of elementary hexahedral "bricks". Each brick (or element) is then transformed into a mass-spring system. Edges are replaced by springs that connect masses representing the vertices. The key point of our approach is the determination of the stiffness springs to reproduce the correct mechanical properties (Young's modulus, Poisson's ratio) of the reconstructed object. We validate our methodology by performing some numerical experiments. Finally, we evaluate the accuracy of our approach, by comparing our results with the deformation obtained by finite element method.
  • Article
    Full-text available
    This paper presents a modeling system which takes advantage of two-dimensional drawing knowledge to design three-dimensional free-form shapes. A set of mouse or tablet strokes is interpreted by the system as defining both a two-dimensional shape boundary and a displacement map. This information is used for pushing or pulling vertices of existing surfaces, or for creating vertices of new surface patches. To relieve the burden of 3D manipulation from the user, patches are automatically positioned in space. The iterative design process alternates a modeling by drawing sequence and a viewpoint change. To stay as close as possible to the traditional drawing experience, the system imposes the minimum number of constraints on the topology of either the strokes set or the resulting surface.