The University of Michigan Medical School (UMMS) and College of Engineering is actively engaged in an institution-wide research activity focused on what we call "Tissue Engineering." Central to this effort is activity in biomedical imaging and visualization at all levels in the spatial hierarchy, from gross anatomical to molecular. To this end, we have been active for several years in developing the technological and intellectual resources necessary to enable this activity, ranging from the creation and networking of a tissue engineering imaging center, a musculoskeletal biomechanics modeling facility, and a University of Michigan high performance microscopic image restoration, modeling and visualization computing core.

In close collaboration with the University of Michigan School of Information (UMSI), we are developing an electronic image archive and retrieval system that is capable of organizing our growing microscopic digital image collection from an anatomical structure point of view, linking medical language resources such as the UMLS and MESH with images of anatomical structure centered around research themes. Some of the motivation for this has been discussed by Rosse [1]. We have chosen to use the visible human as a simple point of entry into this large image repository.

The problem we have is simple to understand. We are generating greater than 2 Terabytes of microscopic 2-D and 3-D research microscopic digital imagery per year at UMMS, being digital 3-D confocal, electron, and bright field light microscopes, as well as film scanning equipment. We have chosen to work with several active research groups to tackle this effort within the imaging core facility. These include two imaging intensive projects with the Kresge Hearing Research Institute (KHRI) [2], a 3-D study of the developing drosophila heart, two kidney tissue imaging projects [3.4], a study examining the 3-D structure of normal and toxically treated seminiferous tubules [5], a long study in muscle denervation and regeneration [6], and a large clinical imaging trial (greater than 1 Terabyte imagery) examining the sural nerve of diabetic and normal patients. These projects are listed in Figure 1 and are schematically related to their anatomical position in the Visible Human Male. This system, which we will demonstrate at the meeting, will allow access to these images simply by clicking a mouse on the appropriate position on iconic visible humans, giving simple access across the networked archive and allowing for real-time comparison to the NLM Visible Human structure as desired by the user.

Figure 2 depicts a schematic representation of an evolving prototype medical research microscopic image library of the inner ear. This model features language and anatomical "descriptor maps" of the inner ear at its core and gives the user the ability to navigate the electronic archive of digital images obtained by the researchers at KHRI by searching the database for image, text, or other experimental objects or parameters. Links to explode into the relevant medical literature are planned. In the fall we expect to generalize this approach to several of the projects listed above, to allow for interactive and comparative 2-D and 3-D anatomical studies using UMMS derived research imagery with the standard Visible Human image data sets.

We gratefully acknowledge research support from the National Library of Medicine (AJW and BDA), the Defense Advanced Research Projects Agency (BDA), the Whitaker Foundation (BDA), and the University of Michigan Office of the Vice President for Research (BDA). This paper is dedicated to the memory of our friend and collaborator, Professor Miranda Lee Pao.

1. Rosse, C. The potential of computerized representations of anatomy in the training of health care providers. Academic Medicine 1995. 70(6):499-505.

2. Raphael Y.; Athey, B.; Yang, W.; Lee M.; and Altschuler, RA. F-actin spectrin, and tubulin in the organ of Corti: Comparative distribution in different cell types in mammalian species. Hearing Research 1994, 76:173-187.

3. Chen, M; Harris, NP; Rose, D; Smart, A; He, X-R; Kretzler, M; Briggs, JP; Schnermann, J; Renin and renin mRNA in proximal tulules of the rat kidney. J. Clin Invest. 94: 237-243, 1994.

4. Garz, R; Weinberg, JM; Ortega-Lopez, J; Davis, JA; Venkatachalam, MA. Conservation of structure in ATP depleted proximal tubules. Role of calcium, and polyphosphoinositides and glycine. American Journal of Physiology. 265: F605-F623, 1993.

5. Welsh, MJ; Wu, W; Parvinen, M; Gilmont, RR. Variation in expression of hsp27 messenger RNA during the cycle of the seminiferous epithelium and co-localization of hsp27 and microfilaments in Sertoli cells of the rat. Biol. Reprod. In press.

6. Carlson, BM. The sattelite cell and muscle regeneration: The degeneration and regeneration cycle. SL Gordon, SJ Clair and LJ Fine (eds). Repetitive Motion Injuries of the Upper Extremity. AM Acad Orthpaed Surg, Rosemont, IL, pp 313-322.