New Directions for CAD
(Bio-CAD as an example)
John M. Switlik

One thing to discuss is Knowledge Technology’s (KT) role in the future of CAD. The most well-known variant of KT has been KBE (Knowledge Based Engineering [1]) which has demonstrated remarkable success in several domains. Recently, KBE underwent radical changes as it adapted to improvements to the computational basis. An example of the change is the evolution of CATIA v5 KBE tool suite.

KT has many variants and advancements, and it may be that a good way to start a discussion of KT issues is to look at CATE (Computer Aided Tissue Engineering [see Ref. 4-6]) and at what it entails in terms of current process and requirements.

For starters, let’s look at two recent NewsNet articles.

• In the first KT article, Kurt Swanson discussed the dichotomy [2] between natural ‘forms and manmade objects’ within the context of CATE.
  
  CATE pertains to modeling of articles ‘that look and function’ similar to the natural structure that they are to replace. That is, CATE deals with designs which have naturally occurring cohorts. For usual CAD situations, this is not the case.
  
  Kurt suggested that one way to design these natural structures would be through a new set of primitives that allow mathematical exploration of relationships, such as between ‘pressure and curvature’ and that search the design space by solving mathematical systems that denote physical characteristics, such as minimal energy.



• The second KT article provided quick review of CATE requirements and shows that these are of a much large order of magnitude, given CATE’s domain of the biologic [3], than for the normal design situation.
  
  As mentioned earlier, CATE has to inter-relate its results with functionally complete parts that are complex. According to Kurt, constraints related to these natural parts do not bode well for processes based upon ‘logical equivalence’, where ‘logical equivalence’ pertains to maintaining ‘truth’ relations between the designed and the physical parts.
  
  These relations, as handled in usual CAD, attempt to manage several things, such as how well the design and the physical part match. Much design and analysis effort, especially as augmented by the KT’s KBE variant, looks to identify and to elaborate these types of relationships. Can we expect that the ‘logical equivalence’ approach will continue to work, given that CAX demands are becoming more complicated through time? This remains to be seen; KT has to be at the forefront of this type of analysis.
  
  Using CATE as the example, we can bring into the discussion other approaches that may be apropos to KT. For instance, Kurt argues that we need to consider another concept (‘computational equivalence’) as perhaps being more useful to the complex design environment than has been considered so far.
  
  The second KT article describes ‘computational equivalence’ as a new foundation for design and analysis. This foundation consists of three things: (1) ‘initial boundaries’, (2) a ‘set of instructions’ (automata), and (3) the ‘environment which constrains and modifies the development’ of the design. In a sense, to notice a ‘genetic’ flavor to this triad would not be in error.
  
  The two types of ‘equivalence’ differ in several ways. The ‘logical’ type presupposes an analytic framework that is supported by experience and knowledge. Some use the term ‘closed-form’ in this context, but essentially the idea is that thought precedes artifact. As mentioned above, this type has been the basis for and has been growing with CAD.
  
  But, as work with non-linear systems and complexity has shown, there needs to be more awareness of the ‘computational equivalence’, of its importance to KT, and of the issues of control. A subsequent article will cover ‘computational equivalence’ and its triad more fully.

For now, in order to more fully understand the CATE-influenced variants, we need to look at what this discipline entails. Wei Sun of Drexel University’s CATE Program [4] has several papers that look at Bio-CAD. An overview paper gives examples of Bio-CAD’s applications in CATE [5] and describes the three main categories of CATE, as follows.

• Computer-Aided Tissues and Bio-Modeling includes ‘Anatomic and Biophysics Modeling’ (familiar topics such as geometry morphology, volumetric representation, and mechanics), ‘CAD-based Modeling’ (contour, surface, and solid processing), and ‘Biomedical Physical Modeling’. This category includes processing related to CT or MRI images using reverse engineering methods which are of interest to CAD [see Ref. 7-9].



• Scaffold Informatics and Biomimetic Design involves ‘Scaffold Informatics & Modeling’ (classification, characterizations, ‘biological intent’), ‘Biomimetic Design’ (multi-constrained design), and ‘Multi-scale Modeling for Biological System’ (multi-scale modeling). The design focus of this category will make use of the above-mentioned triad that will be further discussed in future articles.



• Bio-Manufacturing for Tissue and Organ Regeneration covers ‘bio-manufacturing’ and techniques such as ‘3D Cell and Organ Printing’ that involve driving an artifact from a design.

Wei’s paper on Bio-CAD discusses both roadmaps and processes related to CATE and has plenty examples to show the demands the CATE brings to modeling. Other possible uses of CATE modeling would include simulation, analysis, and surgical planning.

CATE can serve as the basis to discuss KT issues. One example of the relationship can be found in a paper dealing with joining boolean and solid operations [6] to effect a mode that adds ‘reasoning’ to the tool set needed to resolve complex heterogeneous shapes.

The purpose of this article was to introduce CATE and to provide links to further information about the discipline. A subsequent article will apply CATE-influenced metaphors to discussion of KT requirements and its future.

References:
[1] Prasad, Brian “What Distinguishes KBE from Automation”, COE NewsNet, June 2005
[2] Swanson, Kurt “New Directions For CAD #1”, COE NewsNet – October/November 2004
[3] Swanson, Kurt “New Directions For CAD #2”, COE NewsNet – April 2005
[4] Drexel University, CATE, Program for Computer Aided Tissue Engineering
[5] Sun, W., Starly, B., Nam, J. and Darling, A., “Bio-CAD Modeling and Its Applications in Computer-Aided Tissue Engineering”, Computer-Aided Design, Vol. 37 (11), 2005, 1097-1114.
[6] Sun, W. and Hu, X., “Reasoning Boolean Operation Based CAD Modeling for Heterogeneous Objects”, Computer Aided Design, Vol. 34, 2002, pp. 481-488.
[7] Switlik, John M. and Macy, B. “Reverse Engineering – Support for An Empirical Process”, COE NewsNet – September 2003
[8] Macy, Bill and Switlik, J. M. “Reverse Engineering – Acquisition Step”, COE NewsNet – October 2003
[9] Switlik, John M. and Macy, B. “Reverse Engineering – Post-Process Step”, COE NewsNet – January 2004

About the Author
John M. Switlik (john.m.switlik@ieee.org) has had experience in many roles related to computing. His interests cover a broad spectrum: computational support for applied science, mathematics, and management; lifecycle and veracity control for system, application, and domain engineering; integration, use, and quality of computational knowledge.