Developmental Imaging Workshop

Developmental Imaging Workshop

A complete account of the Workshop held on September 18 and 19 1997, including the Agenda, Abstracts, Speaker List, and links to Project Web Pages. The title of each talk is linked to its abstract; the speaker name is linked to their address.

CONTENTS: Introduction; Session I Embryological Collections; Session II Resources on the Web; Session III Computers and Imaging in Embryology Teaching and Training; Session IV Imaging in Embryology Research; Session V Imaging in Diagnosis and Treatment; Session VI Funding Opportunities

Speakers

Links to ten cool projects

Introduction

This Workshop demonstrates the uses of computer-assisted imaging in the field of embryonic and fetal development. The first session deals with existing Collections of Historic Embryonic Material, and the second session deals with Digital Resources on the Internet. The subsequent three sessions deal with uses for imaging, namely for Teaching and Training (Session III), for Research (Session IV), and for Diagnosis and Treatment (Session V). The last session describes the Interests of Several Funding Agencies (Session VI), regarding training and research projects utilizing different imaging technologies.

Back to Table of Contents

Session I: Embryological Collections
Chair: Elaine Young, Comparative Medicine Branch, National Center for Research Resources, NIH
The Human Developmental Anatomy Center; History and Current Initiatives: Adrianne Noe, Director, National Museum of Health and Medicine, Armed Forces Institute of Pathology
The Patten Embryology Research Collection: Alphonse R. Burdi, Director, University of Michigan Medical School
Collecting Human Embryos and Fetuses: Thirty-five Years of Experience: Alan G. Fantel, Department of Pediatrics, University of Washington
Embryo Sections on CD-ROM for Studies of Human Development: Raymond F. Gasser, Director, Computer Imaging Lab, Department of Cell Biology and Anatomy, Louisiana State University Medical Center
Heart Imaging in Human Embryos: Kent L. Thornburg and Jeffrey Pentecost, Congenital Heart Research Center, Oregon Health Sciences University

Back to Table of Contents

Session II: Resources on the Web
Chair: Louise Ramm, Deputy Director, National Center for Research Resources, NIH
Imaging and Distribution of Embryo Images and Models.: Elizabeth C. Lockett, Imaging Specialist, Human Developmental Anatomy Center, National Museum of Health and Medicine, AFIP
Magnetic Resonance Imaging and Volume Rendering for Three-Dimensional Analysis of Embryos.: Bradley R. Smith, Center for InVivo Microscopy, Duke University
Gene Expression Information Resource for Mouse Development: Concept, Current Status and Future Goals.: Martin Ringwald,, Gene Expression Information Resource Project, The Jackson Laboratory

Back to Table of Contents

Session III: Computers and Imaging in Embryology Teaching and Training
Chair: Robert S. Ledley, Medical Computing and Biophysics Division, Georgetown University Medical Center
Multimedia Anatomy Tutorial: Animating Developmental Concepts in Three Dimensions: Carmen L. Arbona, MouseWorks, The Visible Embryo, UC San Francisco
Basic Embryology Review Program, An Effort in Collaborative Development.: Albert O. Shar, Executive Director, Computing and Educational Technology, Basic Embryology Review Program, University of PA School of Medicine
The Muritech Internet Atlas of Mouse Embryology.: B.S. Williams, MuriTech, Inc
Embryo Images: Normal and Abnormal Mammalian Development: Tim Poe, University of North Carolina, Chapel Hill
Computerized 3-D Visualization of Embryonic Development: David Damassa, Envision Development Corporation and Tufts University Schools of Medicine, Dental Medicine and Veterinary Medicine
Moving to a Digital Library for Medical Research and Education: Brian Athey, University of Michigan Medical School
Computer Applications for Fetal Imaging: Ultrasound Education and Training: Wesley Lee, Fetal Imaging Resources, Division of Fetal Imaging, William Beaumont Hospital
The Visible Human Project: A Public Resource for Anatomical Imaging: Michael J. Ackerman, Assistant Director for High Performance Computing and Communications, National Library of Medicine, NIH

Back to Table of Contents

Session IV: Imaging in Embryology Research
Chair: Sally A. Moody, Programs in Neuroscience and in Genetics, George Washington University Medical Center
Development and Application of a 3-Dimensional Dynamic Image Analysis System: David Soll, Department of Biological Sciences, University of Iowa
Using microMR Imaging in Developmental Biology - Fish, Frogs, Mice, & Monkeys: Russell E. Jacobs, Biological Imaging Center, Caltech www.gg.caltech.edu/~rjacobs/rjacobs.html
Analysis and Manipulation of Mouse Embryonic Brain and Heart Development with High Frequency Ultrasound Imaging: Daniel H. Turnbull, Skirball Institute of Biomolecular Medicine, New York University Medical Center
Imaging Techniques For Studying The Embryology Of Epididymal-Testicular Descent: Dale S. Huff, Director, Developmental-Perinatal Pathology, The University Of Pittsburgh And Magee-Womans Hospital
Fast Volume Visualization with T-Vox.: Michele Ursino & Gregory L. Merril, HT Medical
Fitting 3D Imaging Data Using Orthogonal Distance Regression: Janet Rogers, Mathematical and Computational Sciences Division, National Institute of Standards and Technology

Back to Table of Contents

Session V: Imaging in Diagnosis and Treatment
Chair: Felix de la Cruz, Chief, Mental Retardation & Developmental Disabilities Branch, NICHD, NIH
3-Dimensional Ultrasound Imaging of the Fetus: Dolores Pretorius, UCSD 3D Ultrasound Imaging Group, University of California, San Diego
The Impact of Fetal Echocardiography on Early Intervention of Congenital Heart Disease: Ernerio T. Alboliras, Director, Non-Invasive Imaging and Fetal Cardiology, Rush Children's Heart Center
Ultrafast MRI in the Evaluation of the Abnormal Fetus During the Second and Third Trimesters: Anne M. Hubbard, The Department of Radiology and the Center for Fetal Diagnosis and Treatment, The Children's Hospital of Philadelphia
The Current Status and Future Potential of Fetal Intervention - Image is Everything: Alan W. Flake, Director, Children's Institute of Surgical Science, Childrens Hospital of Philadelphia
Quantitative Assessment of Mouse Fetal Cardiac Anatomy by Magnetic Resonance Imaging: Harvey Hensley, Susan Schachtner, Anne Hubbard, John Haselgrove & Scott Baldwin, Childrens Hospital of Philadelphia
Power Doppler in the Assessment of Normal and Abnormal Fetal Vascularity: Beverly G. Coleman, Director of Ultrasound Imaging, University of Pennsylvania Medical Center

Back to Table of Contents

Session VI: Funding Opportunities
Chair: A. Tyl Hewitt, Chief, Developmental Biology Genetics & Teratology Branch, NICHD, NIH
Towards an Information Infrastructure for Healthcare: Funding Opportunities Through The Advanced Technology Program: Bettijoyce Lide, Technology Administration, National Institute of Standards and Technology
NSF support for research and infrastructure in developmental biology: Christopher Platt, Division of Integrative Biology & Neuroscience, National Science Foundation
Support Mechanisms for Technology Development at NCRR: Abraham Levy, Biomedical Technology Branch, National Center for Research Resources, NIH
Funding Mechanisms and Opportunities at the NICHD: Steven Klein, Program Official, Developmental Biology, Genetics & Teratology Branch, NICHD, NIH

ABSTRACTS THE HUMAN DEVELOPMENTAL ANATOMY CENTER, NATIONAL MUSEUM OF HEALTH AND MEDICINE, ARMED FORCES INSTITUTE OF PATHOLOGY: HISTORY AND CURRENT INITIATIVES Adrianne Noe, Director, National Museum of Health and Medicine, Armed Forces Institute of Pathology

www.afip.mil/museum/museum.html This presentation will introduce the origins, present work, and future plans of the Human Developmental Anatomy Center of the National Museum of Health and Medicine, Armed Forces Institute of Pathology. Within the context of an overview of nineteenth and early twentieth century developmental anatomy, the Carnegie Human Embryology Collection's history will be presented and its role in advancing knowledge in the field will be offered. With that collection as its core, the Human Developmental Anatomy Center has amassed a number of other significant collections; they will be identified and described.

Several types of work are on-going in the Center. They fall into several categories: research projects, image distribution and modeling projects, publication projects, conservation and collections management projects, and exhibitions. Examples of each will be described. Future plans for the Center as a national research resource and as a component of the National Museum of Health and Medicine will be discussed, as will its role in the National Health Exhibitions Consortium. Finally, procedures for gaining access to the collections by qualified scholars will be addressed and explained.

THE PATTEN EMBRYOLOGY RESEARCH COLLECTION Alphonse R. Burdi, Curator, Patten Embryological Collection, The University of Michigan Medical School Historical Perspectives. The Patten Embryology Research Collection was established at Michigan in the early 1900s due to the need for accurate descriptions of human embryos at critical developmental stages. While the Carnegie Collection featured a world-renown collection of human embryos at their very youngest stages, it seemed both desirable and strategic for Michigan's collection to focus on the later stages of embryonic, second trimester, and early third trimester human specimens. The University of Michigan Collection is thought to be largest collection of human embryos and fetuses found within the United States today, and is a national resource available to investigators throughout the world.

New Goals and Foci. Since 1972, strategic planning involving past, present, and potential users of the Michigan Embryology Collection pointed toward the essentiality of much-needed medical or teratologic flavor for the Collection whereby newly acquired human embryos and fetuses would be documented with medical and familial histories. Since that time, the acquisition of new specimens (approximately 150-200 per year) has been ongoing due largely to cooperative enterprise between the Department of Anatomy & Cell Biology and obstetricians, pediatric dysmorphologists, and pathologists in the University of Michigan Hospitals. This consortium is known as the University of Michigan Teratology Unit. In addition, our Teratology Unit regularly takes first call on receipt of the specimens from throughout Michigan, conducts the necropsies, and provides reports to the referring physicians. Specimens are subsequently received by the Patten Embryology Collection and accounted for as required by anatomical donation laws.

A Shared Resource. Every effort is made to share the collection specimens and derived information with responsible investigators from within and outside the university. A computerized data base catalog has been designed for the easy retrieval of such information dealing with specimen histories (always handled confidentially) that include vital statistics of each specimen, and maternal and familial histories.

A specimen dossier (coded for confidentiality purposes) is then maintained for each of the specimens collected since 1972 and made available to investigators who wish to assess normal and abnormal human morphogenesis using dependent variables that go beyond such traditionally-used measures as crown-rump and crown-heel lengths. Thus, users of the Michigan Collection from throughout the world are able to study prenatal morphogenesis along population lines as are carried out in animal teratologic studies and clinical studies of human birth defects. Investigators contribute to the dossiers with data from their specific studies so that new information might be shared with other users of the same specimens. Specimens and facilities of the Patten Embryology Collection continue to be available to visiting scientists with formalized research protocols.

The currently available human embryos and fetuses, especially the 3500 plus documented specimens collected since 1972 are chiefly second and third trimester aborti. This group is a part of the 1,750 specimens that are currently available as serially-sectioned specimens. While the Patten Embryology Collection continues with its longstanding focus on human morphogenesis, an adjunct collection of serially-sectioned animal (e.g., chick, rats, mice) embryos and fetuses is also available for comparative studies. Every effort is made to accommodate visiting investigators. Located in the Department of Anatomy & Cell Biology, space and equipment are available for the study of specimens within the research facility.

Researchers are welcomed to work in the Patten Embryology Collection by appointment and are invited to submit to the Collection's director a request to work with the collection. The information should include a description of the specimens needed in terms of age or size, population characteristics, type of sectioning, space and equipment needed, and approximate duration of the proposed visit. Submission of the protocol is requested early enough for purposes of scheduling and advising the prospective investigator on the availability of desired materials.

COLLECTING HUMAN EMBRYOS AND FETUSES: THIRTY-FIVE YEARS OF EXPERIENCE Alan G. Fantel, Department of Pediatrics, University of Washington The Central Laboratory for Human Embryology has collected, studied and analyzed human conceptal tissue for nearly 35 years. During that time, changes in laws, practice, funding and medical technology have significantly affected the condition, types and stages of embryonic and fetal tissues available for research. We will discuss these changing relationships, detailing tissue availability today and present information on obtaining material for biomedical research, including imaging. During the 1960s, most tissue collected by the laboratory was delivered by spontaneous abortion. Despite this designation, the percentage that was actually induced could not be determined. While these specimens enabled studies of abortus morphology and cytogenetics, they were generally too autolyzed for imaging, biochemical or molecular studies. Tissue available for research was largely derived from hysterotomy and hysterectomy and to a lesser extent, from surgical intervention in ectopic pregnancy. Hysterotomy specimens tended to date from mid second through third trimesters and preservation was generally poor. They were often nonviable in utero for days or were exposed to KCl prior to delivery, inducing rapid and severe autolysis. Specimens derived from hysterectomy tended to be in excellent condition but generally dated from relatively late gestation. Ectopic specimens served as a primary source of embryos but massive bleeding and tubal rupture commonly limited successful retrieval. These rare specimens remain an important source of intact, early material.

In the 1970s Washington enacted therapeutic abortion on demand. Most terminations were performed by dilatation and manual curettage and it became possible to obtain increasing amounts of well preserved tissue. Embryos tended to be relatively intact and most organs and tissues could be collected. Increasing use of vacuum extraction in the late 1970s made tissue retrieval challenging, since specimens tended to be severely fragmented by exposure to pressure gradients and passage through fine cannulas, long runs of tubing and collection in gauze bags. Most of these specimens date from late first and early second trimesters with increasing availability of late second and third trimester fetuses. In the late 1980s and early 90s, they were often pretreated with KCl, rendering them useless for detailed study.

Today, the Central Laboratory for Human Embryology supplies university and institute-based investigators nationwide with embryonic tissues processed according to the requirements of individual studies. Technicians collect at clinic sites where specimens are obtained within minutes of passage, rapidly assessed and staged. Individual tissues are then identified, separated, processed and delivered to the laboratory from which they are shipped by overnight air express. Information on obtaining embryonic or fetal tissue can be obtained at 800-583-0668.

Back to Table of Contents

EMBRYO SECTIONS ON CD-ROM FOR STUDIES OF HUMAN DEVELOPMENT Raymond F. Gasser, Director, Computer Imaging Lab, Department of Cell Biology and Anatomy, Louisiana State University Medical Center, New Orleans, Louisiana. The serial sections of human embryos were captured digitally and stored on CD-ROM, making them available to investigators and instructors at minimal expense. All of the sections of three human embryos from the Carnegie Collection were captured at 1024 X 768 pixels (16 or 32 bits/pixel) in targa format. Image processing was used to delete debris, repair tears and folds, color enhance faded stains and sharpen contrast. Each section was placed in alignment, given a scale bar, section number and slide location number, then stored on CD-ROM in TIFF format with LZW compression for distribution. The enhanced sections can be viewed in sequence with most WWW browsers (e.g., Netscape) or any image viewer (e.g., Photoshop) on an inexpensive PC 486 computer. Besides preservation and distribution of valuable reference material, examples will be presented on how the data can be utilized by investigators (determination of 3D growth movements from reconstructions) and instructors (construction of interactive atlases of sectional morphology).

HEART IMAGING IN HUMAN EMBRYOS Kent L. Thornburg and Jeffrey Pentecost, Congenital Heart Research Center, Oregon Health Sciences University The Congenital Heart Research Center (CHRC) at Oregon Health Sciences University has directed construction of 3D computer models of embryonic human hearts with the intention of 1) devising a mechanism for researchers to store, retrieve, and visualize physiologic, molecular, and genetic data pertinent to cardiogenesis, and 2) improving methods for teaching normal/abnormal cardiac embryology. Using serial sections of human embryos from the Carnegie Collection at the Armed Forces Institute of Pathology, contours are manually traced, then re-stacked into three-dimensional, wire frame models. A surface connecting the layers provides the appearance of a solid model.

Because of its versatility and ease of manipulation, the resultant structure is an ideal tool for visualizing structural and experimental data. For example, 3D simulations of embryologic hemo-dynamics and internal/external heart development based on real data provide a new means of analyzing influences of mechanical forces on cardiogenesis. Consequences of genetic expression affecting heart formation can be mapped in these models, showing exactly which tissues are involved at specific stages of development. As an educational tool, this system of viewing the embryonic heart in all of its developmental stages surpasses any currently used method. The user can view the developing heart in cross-section from any angle, observe 'morphing' from stage to stage, and 'fly' through the computerized model.

Future Plans
Human model construction: The CHRC is currently collecting more data from the Carnegie Collection of Embryos; by the end of the summer of 1997, the CHRC plans to have data for six additional embryos of different developmental stages.

Mouse model construction: An image database is being developed for similar reconstruction of the embryonic mouse heart. The process of specimen preparation, capturing digital photomicrographs, transferring images to the computer, and tissue segmentation for reconstruction has been established at CHRC. Modeling of congenitally malformed mouse hearts is one of our next steps. An examination of the influences of hemodynamic forces and blood flow patterns on cardiogenesis is being planned. This project addresses one of the most exciting new theories of heart development, and will depend on the physical heart models created by the CHRC.

Stereolithography: CHRC computerized heart models are being converted to physical models using a process by which computer data directs the production of physical models using laser beams and resin-based materials. The accuracy and precision of such a process is exceptional. This technology has not been applied previously to anatomical modeling at this level. The National Museum of Health and Medicine in Washington, DC, has requested our first model for permanent display in their collection. These physical models will be used by CHRC scientists for analysis of embryonic hemodynamics. They will also serve as an invaluable teaching tool for medical and scientific education.

Back to Table of Contents

IMAGING AND DISTRIBUTION OF EMBRYO MODELS FROM THE HUMAN DEVELOPMENTAL ANATOMY CENTER, NATIONAL MUSEUM OF HEALTH AND MEDICINE, ARMED FORCES INSTITUTE OF PATHOLOGY E. C. Lockett, Human Developmental Anatomy Center, National Museum of Health and Medicine, Armed Forces Institute of Pathology

magenta.afip.mil/embryo/ This presentation will discuss current strategies for imaging histologic information from the Carnegie Collection, modeling techniques, and distribution mechanisms for both models and histologic information. A re-evaluation of the target audience for models and images prompted by recent challenges to distribution methods has required changes in storage and distribution methods. Long term logistical goals for the distribution of information from the Center will be discussed, as well as steps taken toward achieving these goals.

MAGNETIC RESONANCE IMAGING AND VOLUME RENDERING FOR THREE-DIMENSIONAL ANALYSIS OF EMBRYOS

Bradley R. Smith, Department of Radiology, Duke University, Durham, NC

wwwcivm.mc.duke.edu/civmPeople/SmithBR/documents/HumanEmbryo The difficulty in obtaining well-documented and well-preserved human embryo specimens presents an important challenge to understanding normal and abnormal development. There is a need to minimize the number of embryos of all species used during research and to maximize the distribution of information obtained from the embryos that are used. A complete source of distributable, three dimensional image data representing the human embryological time period is not yet available. The Multidimensional Human Embryo project, will generate three-dimensional image data sets representing most of the human embryonic time period, and will make these data easily accessible. The data will be generated by magnetic resonance imaging with specimens from the Carnegie Collection of Human Embryos. Embryos from stage 12 through stage 23 will be scanned three times each to produce T1, T2, and Diffusion weighted data sets. Each data set will be isotropic with 256 x 256 x 256 voxels. The image data will be made available at the conclusion of the project by CD-ROM and via a Web site. Sample images will also be available via the web site throughout the project to document progress of the work.

Magnetic resonance imaging is also being used to characterize the three dimensional structure of normal and abnormal animal embryonic hearts. The vasculature is represented as three-dimensional digital models that can be measured, rotated to any orientation, and digitally sectioned to permit rapid comparison of normal and abnormal hearts. Genetically manipulated mouse embryos (gap junction gene Cx43 knockouts and CMV43 overexpression) and surgically altered chick embryos were investigated by MR microscopy after vascular infusion with the MR contrast agent Gd-DTPA-BSA. Blood flow was altered in chick embryos by partial left atrial ligation with nylon suture. MR imaging of the mouse and chick embryos was performed at 9.4 T to produce 128 image slices each with 2562 pixels at 54.7µm resolution. Three-dimensional spin warp encoding produced T1-weighted datasets in 3.5 hours per embryo. MR imaging demonstrated the morphological changes in the right ventricle, conotruncus, and ductus arteriosus associated with altered expression of the Cx43 gap junction gene in mouse embryos. It also demonstrated the altered aortic arch morphogenesis due to the changing blood flow pattern in the left atrial ligated chick.

GENE EXPRESSION INFORMATION RESOURCE FOR MOUSE DEVELOPMENT: CONCEPT, CURRENT STATUS AND FUTURE GOALS M. Ringwald, R. Baldock*, J. Bard#, D. Begley, G. Davis, D. Davidson*, C. Dubreuil*, J.T. Eppig, K. Frazer, P. Johnson, M. Kaufman#, M. Mangan, J. Richardson, L. Trepanier. The Jackson Laboratory, Bar Harbor; *MRC Human Genetics Unit, Edinburgh; #Edinburgh University, Edinburgh.

www.informatics.jax.org/doc/gxdgen.html The process of differential gene expression generates extraordinarily complex spatio-temporal networks of gene and protein interactions. A major thrust of current biomedical research is elucidating these networks to understand the molecular basis of human development, health, and disease. The laboratory mouse serves as a pivotal animal model in these studies. To cope with the shear volume and complexity of gene expression data, we, the Gene Expression Database (GXD) group at The Jackson Laboratory and our collaborators at the MRC Human Genetics Unit and the University in Edinburgh, Scotland, are developing a comprehensive Gene Expression Information Resource for Mouse Development. The resource will link the following components:

(1) GXD, which stores and integrates many types of expression data and provides comprehensive links to the Mouse Genome Database (MGD) and to other databases containing DNA and protein sequence information, genetic and physical mapping data, and descriptions of disease states and mutant mice to place the gene expression data into the larger biological and analytical context. GXD describes expression patterns by a controlled anatomical dictionary that is part of the Anatomy Database and includes 2D images of original in situ data that are indexed via terms from the dictionary.

(2) The Anatomy Database, which provides the standard nomenclature for textual queries relating gene expression to developmental anatomy.

(3) the 3D Atlas of Mouse Development, high resolution digital 3D models of mouse embryos at representative developmental stages reconstructed from serial sections, which enable 3D graphical storage, display and analysis of expression patterns. The major anatomical structures in the 3D atlas are labeled using the anatomy nomenclature. This assignment provides an important link between the graphical and the text based query systems.

The Gene Expression Information Resource will provide the research community with a tool to store and analyze gene expression data in the appropriate context using the full power of combined text and image-based methods. The current status of the project will be presented and future applications of the resource for biomedical research will be discussed.

The GXD Project is supported by NIH grant HD33745. Work at Edinburgh is supported by the MRC, the BBSRC, and by the European Science Foundation.

Back to Table of Contents

MULTIMEDIA ANATOMY TUTORIAL: ANIMATING DEVELOPMENTAL CONCEPTS IN THREE DIMENSIONS Carmen Arbona, MouseWorks, San Francisco, CA

visembryo.ucsf.edu The Visible Embryo (http://visembryo.ucsf.edu) is an interactive early anatomy program for medical student training delivered via the internet as a project of the Multimedia Anatomy Tutorial (MAT). The Visible Embryo illustrates the first four weeks of in-utero development and is being expanded to illustrate organ development in 3D specifically 1) the processes of gastrulation and neurulation, 2) development of the heart tube illustrating congenital heart defects, and 3) the formation of organ systems. The Visible Embryo program is based on evidence that visualization in 3D assists in the understanding of difficult concepts. A correct understanding of the early embryonic concepts is essential to medical education and practice and useful to a lay audience in understanding infertility and birth defects. MAT is an outcome-based teaching curriculum developing delivery of medical education on-line.

BASIC EMBRYOLOGY REVIEW PROGRAM (BERP), AN EFFORT IN COLLABORATIVE DEVELOPMENT Albert O. Shar, Computing and Educational Technology, University of Pennsylvania School of Medicine

www.med.upenn.edu/meded/public/berp In 1990, the University of Pennsylvania School of Medicine had a well developed program of medical software development funded by a contribution of hardware from Apple Computer and for programming from the Pew Charitable Trust. The program was based upon the use of technical staff for the development of core software (programs, templates, training, special purpose equipment, standards, etc.), medical students to provide the actual combination of text and images and limited programming as needed and faculty as the content experts. One of the most successful programs developed under this effort was BERP(c), the Basic Embryology Review Program. This Macintosh-only program has been extensively used at the U of P and other locations.

Shortly after this program was developed, funding for education software development became essentially unavailable. By 1995, the power of the World Wide Web as a method for delivering, platform independent multimedia was becoming clear and, as a proof of concept, an attempt at moving BERP to the web was started. This still unfinished and relatively naive program was largely ignored for two years, with only occasional attempts to refine the product. Even with this minimal attention, this application is consistently among the most popular areas visited on the entire UPHS web.

In May of 1997, in part because of the funding for a new curriculum, funds were made available to refine and complete the web version of BERP(c). This effort is scheduled for completion by the end of September 1997.

THE MURITECH INTERNET ATLAS OF MOUSE EMBRYOLOGY B.S. Williams*, M.J. Pescitelli,* and M.D. Doyle*^# MuriTech Inc., Cambridge, MA* and Eolas Technologies Inc., Chicago, IL^#

www.muritech.com/

www.eolas.com The Muritech Internet Atlas of Mouse Embryology allows visualization of the anatomy of the mouse embryo with linkage to information about the developing structures. The system uses the resources of the World Wide Web. It provides educators, students and researchers with an easy-to-access interface to an interactive online reference system that correlates textual information with 2-dimensional and 3-dimensional microscopic images of mouse embryos at various stages. These images were obtained by digitizing paraffin serial sections of mouse embryos. The high-resolution multidimensional image data were precisely mapped using the patented MetaMAP(R) Web imagemap system, allowing standard Web browsers to be used to view the images and interactively query the knowledge base. By clicking on any part of the mapped image datasets, relevant information can be retrieved and viewed in an adjacent frame of the Web browser. An applet-based 3-D imagemap browser is also demonstrated that allows the user to rotate and slice through an embryo image dataset, and query the knowledge base by clicking directly on voxels in the 3-D dataset from within the Web browser page.

EMBRYO IMAGES: NORMAL AND ABNORMAL MAMMALIAN DEVELOPMENT Kathleen K. Sulik, Peter Bream, and Tim Poe, The University of North Carolina, Chapel Hill, North Carolina 27599-7090 Embryo Images: Normal and Abnormal Development is an interactive tutorial based on scanning electron micrographs (SEMs) of mammalian embryos (mouse and human). The original CD-ROM has been available since 1994 and is currently in use in the majority of the medical schools in this country. The program has enjoyed popularity, in part, because the SEMs provide a three dimensional-like image of actual embryos. Embryo Images was created using Macromedia's Director. Director is the industry standard for multimedia authoring, allowing products to be ported to multiple platforms and the Internet. Text and graphics have been carefully combined in a manner that is both interactive and easy to use. Digitized movies, animations, and morphs (in this case, movies simulating embryo growth) have been added to illustrate key concepts. Removable semi-transparent color overlays are often included to identify embryo anatomy and a "Help" section is available to introduce users to the features of Embryo Images. We are completing revisions and additions to this program, with major emphasis being placed on inclusion of concepts and illustrations related to abnormal development. The student has the option of looping out at indicated points from study of the sequence of normal embryonic development to view abnormal sequences and consequences. Abnormal sequences of development are typically illustrated in scanning electron micrographs and consequences are typically color patient photographs. Music has been added as a user option for the more narrative components of Embryo Images. The program can be used as an independent resource or as a complement to embryology textbooks. Although the target audience is largely medical students, it is also a valuable resource for developmental biologists and toxicologists, biology students and clinicians. The CD-ROM will run on both Windows '95 and the Mac OS.

COMPUTERIZED 3-D VISUALIZATION OF EMBRYONIC DEVELOPMENT R.L.Simon, D.A. Damassa, R.F. Willson, A.W. Gustafson, and E.W. Overstrom. Envision Development Corporation, Marlboro, MA and Tufts University Schools of Medicine, Dental Medicine and Veterinary Medicine, Boston, MA The overall goal of this project is to generate a database of 3-D digital embryonic images of pig development and, using this database, create low-cost software products for the study of mammalian embryology. Applications are being tailored for medical, dental and veterinary curricula. Reconstructed 3-D datasets of pig embryos are generated from serial digital images obtained by confocal and conventional microscopy. From these datasets, both three-dimensional and slice views of the embryo are produced. A Silicon Graphics Indy workstation running InterVision's 3-D software is used to generate specific views and also produce video clips of rotations of the 3-D images. The ultimate design of these software products will allow users to visualize 3-D images of embryos, view the formation of specific organs and organ systems, and access related hyper-media, including text, histological sections and clinical cases. Key features in product design include direct access to the various databases for customization of program content and network/web-based implementation. MOVING TO A DIGITAL LIBRARY FOR MEDICAL RESEARCH AND EDUCATION Brian Athey, University of Michigan Medical School Not Available

COMPUTER APPLICATIONS FOR FETAL ULTRASOUND TRAINING. Wesley Lee, Fetal Imaging Resources, Inc

www.fetus.com Birth defects are a leading cause for infant mortality in the United States. Prenatal detection can improve outcome in selected cases but, some studies suggest that improved diagnostic training is necessary to improve the detection rate of fetal anomalies. Fetal Imaging Resources, Inc. is collaborating with the American College of Obstetricians and Gynecology and the Acuson Corporation for developing a CD-ROM product that addresses this need.

The multimedia-based software prototype will allow the user to learn about fetal ultrasound anomalies in a Research Library. Diagnostic skills can then be tested through an interactive learning environment called the Examination Room. Advanced users may take the Ultrasound Challenge where a variety of problem-oriented scenarios are presented. User responses will be tracked by time and accuracy.

The presentation will include a demonstration of this prototype product and discussion about how interactive multimedia can offer advantages for prenatal ultrasound training. Possible multimedia delivery scenarios across the Internet will also be examined.

Another project involves simulation of embryonic heart development by three-dimensional animation. Developmental sequences were created with the assistance of Dr. Keith Moore, Professor Emeritus from the University of Toronto. Examples of conotruncal abnormalities (transposition of the great arteries, double outlet right ventricle, and tetralogy of Fallot) will be demonstrated with Quicktime VR technology. This technique allows the user to completely rotate a virtual fetal heart model for close inspection and comparison with digital ultrasound video examples.

Finally, the potential role of three-dimensional fetal ultrasound for medical education will be discussed. Examples of fetal volume reconstructions will be presented to illustrate the application of this technology for prenatal ultrasound training.

THE VISIBLE HUMAN PROJECT^TM: A PUBLIC RESOURCE FOR ANATOMICAL IMAGING Michael J. Ackerman, National Library of Medicine, Office of High Performance Computing and Communications

www.nlm.nih.gov/research/visible/visible_human.html The National Library of Medicine (NLM) has long been a world leader in the archiving and distribution of the print-based images of biology and medicine. NLM has also been a pioneer in the use of computer systems to encode and distribute textual knowledge of the life sciences. NLM's Long Range Planning effort of 1985-86 foresaw a coming era where NLM's Bibliographic and factual database services would be complemented by libraries of digital images, distributed over high speed computer networks and by high capacity physical media. The NLM Planning Panel on Electronic Imaging recommended that NLM should undertake the building of a digital image library consisting of computer assisted tomography (CAT), magnetic resonance imaging (MRI), and cryosection images of a representative, carefully selected and prepared male and female cadaver -- the "Visible Human Project^TM." The male Visible Human data set became available on November 28, 1994. The Female Visible Human data set became available one year later. The Visible Human data sets are being made available through a license agreement with the NLM.

The Visible Human Project data sets are designed to serve as a common reference point for the study of human anatomy, as a set of common public domain data for testing medical imaging algorithms, and as a test bed and model for the construction of image libraries that can be accessed through networks. The data sets are being applied to a wide range of educational, diagnostic, treatment planning, virtual reality, artistic, mathematical and industrial uses by over 800 licensees in 26 countries.

The data sets are having their greatest effect on health care and health education. They are used as a normal reference and as an aid in the diagnostic process. Programs under development will be used to educate patients about the need for and purpose of surgery and other medical procedures as well as to permit physicians to plan surgery and radiation therapy. The images from the Visible Human data sets are used in several prototype virtual reality surgical simulators. Educational materials that make use of the Visible Human data sets are beginning to be used by students from kindergarten to practicing health care professionals.

But key issues remain in the development of methods to link such image data to symbolic text-based data comprised of names, hierarchies, principles and theories. Standards do not currently exist for such linkages. Generalizable methods like the use of hypermedia where words can be used to find pictures and pictures can be used as an index into relevant text are being experimented with. Basic research is needed in the description and representation of structures, and the connection of structural-anatomical to functional-physiological knowledge. This is the larger, long-term goal of the Visible Human Project: to transparently link the print library of functional-physiological knowledge with the image library of structural-anatomical knowledge into one unified resource of health information.

Back to Table of Contents

DEVELOPMENT AND APPLICATION OF A 3-DIMENSIONAL DYNAMIC IMAGE ANALYSIS SYSTEM (3D-DIAS) David R. Soll, Department of Biological Sciences, University of Iowa Because we traditionally view living cells through microscopes in 2 dimensions as they move along a flat substratum, we tend to conceptualize both their behavior and organization in 2 dimensions. In reality, however, both a cell's behavior in vivo and in vitro, and a cell's organization is 3-dimensional. We have, therefore, developed the 3- dimensional Dynamic Image Analysis System (3D-DIAS) to study the behavior of a cell in 3 dimensions. In this system, a migrating cell is optically sectioned through differential interference contrast optics during a period short enough so that cell movement does not cause a reconstruction artifact between the first and last section. Optical sectioning of the entire cell is repeated every second. The optical sections at each time point are then digitized into the 3D-DIAS database, image processed, and the edges of the cell image detected by an invention referred to as the "pixel complexity measurement". Each continuous edge is then converted into a beta-spline model, and filled with the in-focus portion of the original image. This results in the subtraction of all out of focus and extracellular information in the original optical section. The optical sections at each time point are stacked and pixel intensity averaged between optical sections in the Z-axis. This results in the genesis of a complete 3D digitized image every second. 3D-DIAS software then allows computer-generated movies to be produced in which a crawling cell can be viewed by means of a 3D workstation at any angle. Because the 3D-surface of the cell is converted to a beta-spline model, 140 parameters of motility, based on the 3D path of the cell centroid, and of dynamic cell morphology, based on the 3D encapsulating surface, can be computed at one-second intervals. The cell can also be peeled or gouged at any angle to any depth, and the dynamics of the nucleus and intracellular organelles viewed and quantitated. The system also color codes the particulate and nonparticulate cytoplasm of a crawling cell, which is useful in monitoring the 3D dynamics of F-actin filled pseudopods. 3D-DIAS has been applied to a number of developmentally relevant systems, most notably the chemotaxis and aggregation processes in the cellular slime mold Dictyostelium discoideum and mitosis in fibroblast cell lines. In both cases, the power of the system has been in describing aberrant behavioral phenotypes of specific cytoskeletal mutants. 3D-DIAS has been underutilized primarily because it has been continuously under development at Iowa during the last six years and no version has been commercialized. Its potential application to embryogenesis and many other aspects of cell motility are self-evident. A second-generation 3D-DIAS system is now under development that will include near-real time reconstruction speeds, high-speed (250 frame per second) analysis of intracellular vesicles, a confocal front end and virtual reality software. Through funds from the W.M Keck Foundation, visitors can now access 3D-DIAS.

Soll, D.R. 1995. The use of computers in understanding how cells crawl. Int. Rev. of Cytology, 163, 43-104.

Shutt, D., Wessels, D., Chandrasekhar, A., Luna, B., Hitt, A. and Soll, D.R. 1995. Ponticulin plays a role in the spatial stabilization of pseudopods. J. Cell Biol., 131, 1495-1506.

Wessels, D., Titus, M., and Soll, D.R. 1996. A Dictyostelium myosin I plays a crucial role in regulating the requency of pseudopods formed on the substratum. Cell Motil. Cytoskel., 33, 64-79.

Soll, D.R. and Voss, E. 1997. Two and three dimensional computer systems for analyzing how cells crawl. In "Motion Analysis of Living Cells", ed. D.R. Soll, D. Wessels. John Wiley, Inc. In press.

USING mMR IMAGING IN DEVELOPMENTAL BIOLOGY - FISH, FROGS, MICE, & MONKEYS

Russell E. Jacobs, Biological Imaging Center, California Institute of Technology

www.gg.caltech.edu/~rjacobs/rjacobs.html This work is a collaboration among a number of disciplines at Caltech, including neuroscience, developmental biology, computer science, chemistry, and electrical engineering. The long-term goal of this work is to refine and apply a goal-directed interactive paradigm for data collection and analysis in the neurosciences. We are applying it to Magnetic Resonance Micro-Imaging in several contexts: the generation of in vivo atlases of brain development; studies of transgenic mice model systems; and the design of new MR contrast agents. The specific aim of the computational aspect of this work is to create a "teleological pipeline" for making datasets, models, and images. By "teleological pipeline" we mean a computational and instrumental device in which the user's goals (i.e. the qualities they wish the final image to possess) guide the data collection and processing procedures.

Preliminary results obtain using our current 11.7T vertical bore MRI instrument will be used to illustrate this melding of disciplines. We focus on five topics using : MRI as a methodology to examine embryonic development in vivo and in vitro in small animals:

� mMRI of developing frog embryos
� in vitro and in vivo : MRI of mouse embryos
� in vitro : MRI of a small primate (Microcebus murimus)
� diffusion tensor MR imaging of a transgenic MS mouse model system
� bi-functional & 'smart' MR contrast agents We begin with a discussion of 3 dimensional MR imaging of fixed mouse specimens which demonstrates the excellent image quality, resolution, and contrast that can be obtained on immobile samples in a several hour experiment. Preliminary diffusion tensor imaging of a transgenic mouse model for multiple sclerosis illustrates the wealth of information inherent in this imaging modality and its ability to delineate nerve tracts and abnormalities in the mouse spinal cord. Finally, we show some recent in vivo MR and fluorescence images demonstrating the feasibility of employing such bi-functional agents in the developmental studies.

This work was funded by the Human Brain Project with contributions from the National Institute of Drug Abuse, the National Institute of Mental Health, and the National Science Foundation; by the National Institute of Child Health & Human Development; and by the Beckman Institute.

ANALYSIS AND MANIPULATION OF MOUSE EMBRYONIC BRAIN AND HEART DEVELOPMENT WITH HIGH FREQUENCY ULTRASOUND IMAGING Daniel H. Turnbull, Skirball Institute of Biomolecular Medicine New York University Medical Center The availability of genetic analysis and transgenic techniques in the mouse have led to its widespread acceptance as the preferred animal model for studying mammalian development and many human diseases. We have developed a high frequency (40-100 MHz) ultrasound backscatter microscope (UBM), capable of high resolution (~50µm) noninvasive imaging of mouse embryos, in utero. The most obvious features in the real time UBM images of early mouse embryos are the beating heart and the fluid-filled neural tube cavity, which appears echo-free in contrast to the surrounding embryonic tissues. Both of these features have been utilized to perform in vivo analysis of early embryonic brain and heart development.

Cardiac activity can be identified with UBM as early as gestational age 8.5 days (E8.5, equivalent to human day 21). By E9.5 the common atrium and ventricle can be identified, and cardiac chamber dimensions can be traced on UBM images through the early stages of morphogenesis and ventricular septation (E10.5-13.5). UBM imaging is also useful for cardiac imaging of neonatal mice, and has recently been used to characterize functional differences between normal and a-myosin heavy chain (Arg403Gln) mutant mice which die of congestive heart failure in the first week of life. At the high ultrasound frequencies used in UBM imaging, moving blood produces a high signal. As a result, blood flow through umbilical, vitteline and other major blood vessels can be easily delineated as a hyperechoic streaming pattern. Quantitative data on heart rate and blood flow velocities in the embryonic heart and umbilical vessels are collected using a 40 MHz Doppler ultrasound system, recently implemented on the UBM. The ultimate goal of this project is to develop noninvasive methods to assess cardiovascular function in normal and mutant mouse embryos. We have shown that cardiac defects associated with a null mutation of VCAM-1 can be identified in utero with UBM imaging, and are currently quantifying umbilical Doppler waveforms in order to characterize putative placental defects in the same mutant embryos.

The shape of the neural tube cavity, evident from in utero UBM images can be used to detect and analyze neural tube defects such as the mid-hindbrain deletions associated with null mutations of the Wnt-1 and En-1 genes. Three-dimensional UBM images have been used to characterize volumetric differences in the neural tube endoluminal space between E10.5 mutant embryos and normal littermates. In addition, specific brain structures are readily identified with UBM, especially those which appear in contrast to the venticular fluid, including the forebrain basal ganglion and septum, thalamus, midbrain tectum and cerebellum. Similarly, non-neural structures such as the developing limbs can be followed in utero, and the initial formation of digits visualized with UBM.

We have recently modified the UBM system to allow cells, retroviruses and other agents to be injected into early mouse embryos, in utero. The ability to manipulate embryos through transplantation and injections has been widely used in lower vertebrate species such as frog, zebrafish and chick, but has been difficult or impossible in mammalian embryos due to their inaccessibility, enclosed in the maternal uterus. We have used UBM as a guidance system, allowing image-directed, targeted injections into the embryonic neural tube as early as E9.5, and into specific parenchymal target regions in the developing limbs and brain as early as E11.5. The UBM-guided embryo injection system is being used in a number of projects: using retroviruses to study cell lineages in the mouse forebrain and limb, transplantation of neural cells to determine the developmental potential and fate of cells placed in ectopic regions of the brain, and the effects of gene misexpression using both retroviruses and transfected cell lines expressing secreted proteins. The availability of mouse mutants opens up the possibility to study cell lineage and fate and the effects of altered gene expression using UBM-guided injections in mice lacking specific genes, as well as normal animals. In particular, it should be possible to test cell replacement and gene therapy strategies by injecting cells or retroviruses expressing specific genes into defined mutant embryos.

In summary, UBM can be used to analyze noninvasively a variety of developmental processes, and to manipulate mouse embryos through UBM-guided injections over a wide range of embryonic stages. The combination of recent breakthroughs in mouse genetics together with this new high resolution ultrasound imaging technology is providing a powerful system for studying mammalian embryogenesis and human disease models in the mouse.

IMAGING TECHNIQUES, OLD AND NEW, IN THE STUDY OF EPIDIDYMAL-TESTICULAR DESCENT IN HUMAN EMBRYOS Dale S. Huff, MD, Magee-Womens Hospital and the University of Pittsburgh; Elizabeth Lockett, William Discher, and Adrianne Noe, PhD, The Human Developmental Anatomy Center of the National Museum of Health and Medicine, Armed Forces Institute of Pathology. In 1912, Felix cited overwhelming embryological evidence against the universally accepted concept of transabdominal testicular descent. He apparently had a talent, which other embryologists lacked, for reconstructing in his mind an accurate three-dimensional image of an embryo from two-dimensional serial microscopic sections and graphic reconstructions of that embryo and further for reconstructing in his mind an accurate four-dimensional movie loop-like visual image from multiple mental three-dimensional visual images of a series of progressively more mature embryos. In the century since Felix first rejected the concept of transabdominal testicular descent, multiple imaging techniques applied to human embryos have confirmed his position. These techniques have included high quality serial microscopic sections from many more well preserved human embryos, detailed graphic reconstructions, models of whole embryos and of organ systems, photographs of gross dissections, stereo photographs both gross and microscopic, scanning electromicrography, and transmission electromicrography. Despite a century of solid imaging evidence to the contrary, the concept of transabdominal testicular descent in humans has persisted and has recently been used as the foundation upon which theoretical strategies for the treatment of cryptorchidism have been based. Only in the last few years have the modern imaging technologies of computer-assisted three-dimensional reconstruction and morphing provided the potential for reproducing on computer, video, and movie screens what Felix probably saw in his mind�s eye one hundred year ago. These reconstruction and morphing techniques are being applied to the historic Carnegie Collection of Embryology in an effort to confirm Felix�s position. To date, the mesonephros of three embryos, Carnegie Collection numbers 5072 (Stage II, 17 somites), 6097 (Stage 12, 25 somites) and 1380 (Stage 14) have been completed. The results support the concepts that the development of the mesonephros begins and is completed during the fourth post-ovulatory week as part of the cranio-caudal developmental gradient of blastogenesis now known to be under the control of homeobox genes; that the relationship between what will become the tail of the epididymis and what will become the inguinal ring is established by the 28th post-ovulatory day; and that transabdominal testicular descent does not occur in humans. It is hoped that reconstructions of the mesonephros from a complete series of embryos will conclusively demonstrate that Felix was correct in his rejection of transabdominal testicular descent in humans.

Back to Table of Contents

FAST VOLUME VISUALIZATION WITH T-VOX Michele Ursino, Ph.D. and Gregory L. Merril, HT Medical, Rockville, Maryland

www.ht.com Embryonic development is a dynamic, three-dimensional process, posing significant challenges for the research and teaching communities that disseminate current research findings. We have developed software, called T-Vox (Teleos Voxel Visualizer), that facilitates the transformation of 2D data into 3D models and the manipulation of 3D models for the visualization of dynamic processes that occur during embryogenesis. This has involved the development of specific software features, including volume rendering and interactive look-up table manipulation through five editors to isolate bones, skin, and other tissues. T-Vox also can generate animation files in MPEG or QuickTime formats. This feature pre-renders complex voxel images for distribution on PCs and Macintosh computers.

To perform volume rendering with update speeds from 3 to over 30 frames per second, T-Vox uses a special technique based on hardware capable of managing volumetric textures. A volumetric texture is a region where texture values, called texels, are stored in a 3-D grid in a particular position and orientation in space. When the volumetric texture is active, each polygon inside the texture space assumes the color of the texture values. For example, when T-Vox loads data in the texture space of a set of images and draws a polygon through it, users can obtain an arbitrary section of the data. When many polygons are drawn through the texture space and the alpha blending, or transparence effect, is applied, the single polygons disappear and the volume becomes visible. This technology provides a powerful interactive algorithm to display volume data and enables fast rendering speeds.

FITTING 3D IMAGING DATA USING ORTHOGONAL DISTANCE REGRESSION Janet Rogers, Mathematical and Computational Sciences Division, National Institute of Standards and Technology Orthogonal Distance Regression methods provide meaningful least squares estimates for distance calibration problems such as are required for biomedical image processing. A major characteristic of these problems is the prevalence of non-uniformly distributed noise in the data to be fitted. A NIST developed software package, ODRPACK, specifically designed for these circumstances, has been successfully employed to identify parameters and characteristics of such biotechnology models.

This talk will discuss how Image Guided Technologies (IGT) in Boulder, CO, used ODRPACK to locate significant calibration discrepancies within their coordinate measuring machine. This instrument is employed in the design and manufacture of 3D optical localizers that enable surgeons to track the location of a probe inserted into a patient's skull. The development of novel optical property models was efficiently facilitated by application of ODRPACK, and accuracy testing and certification of new IGT systems are presently performed using procedures based on ODRPACK. Some specific features of ODRPACK that make it especially relevant for high precision measurements and uniquely well-suited for modeling 3D image data will be described.

THREE-DIMENSIONAL ULTRASOUND OF THE FETUS Dolores H. Pretorius, and Thomas R. Nelson, Department of Radiology, University of California, San Diego, California

tanya.ucsd.edu/gallery.html Interest in 3DUS has been building as equipment performance and user experience has increased. Scientific papers are being presented at numerous national and international scientific meetings describing clinical results imaging specific organs, trials of new equipment and determination of optimal methods of acquisition and interactive display of volume. 3DUS imaging can be performed using commercially available ultrasound equipment or in combination with clinical scanners and additional data acquisition and graphics workstations. Since patient imaging geometry often precludes obtaining the optimal image plane, a key feature of 3DUS which is essential to clinical utility, is that the diagnostician may review the volume data interactively, evaluating the relevant anatomy from orientations besides those used to acquire the data. Such flexibility provides an increased understanding of underlying anatomic relationships. 3DUS evaluation of fetuses from various angles permits complex anatomic relationships to be more easily understood. Inclusion of color Doppler imaging facilitates identification of vascular anatomy leading to direct visualization of vessel organization and spatial relationships of the fetus, cord, and placenta. We have shown that volume rendering of fetal structures such as the face provides a more understandable picture of anatomy and that curved bony structures such as the skull and spine can be evaluated in more detail than with 2DUS due to the inclusion of the entire structure rather than a single plane. Interactive display methods are essential to provide the physician with the means to observe and evaluate patient anatomy. Several techniques are used to enhance comprehension, including volume rendering, rotation of volume data, viewing in a standardized orientation and stereoscopic viewing.

The feasibility of 3DUS in the clinical setting is nearby. Benefits of 3DUS include allowing the physician to evaluate arbitrary planes not available with 2DUS due to patient body habitus or fetal position; measure organ dimensions and volumes; obtain anatomic and blood flow information; improve assessment of complex anatomic anomalies; confirm normalcy; standardize the ultrasound exam procedures; enhance understanding of physicians in primary care facilities and communicate volume data over networks for consultation at tertiary facilities. Standardization of the ultrasound examination protocols potentially can lead to uniformly high quality examinations and decreased health care costs.

ADVANCES IN DIAGNOSING FETAL CARDIAC DISEASES BY ULTRASONOGRAPHY Ernerio T. Alboliras, M.D. and Anthony F. Cutilletta, M.D. Rush Children�s Heart Center Rush-Presbyterian-St. Luke�s Medical Center, Chicago, IL. Advancing expertise in obstetric ultrasound techniques allowed diagnosis of an increasing array of fetal anatomic defects, and has resulted in an interdependent relationship between obstetrician/perinatologist and pediatric/fetal cardiologist. During the past decade, major advances in pediatric cardiology have occurred, including a greater understanding of pathology and natural history of congenital heart disease, superior results in neonatal heart surgery, and finer imaging capability, particularly with color flow Doppler and high-resolution two-dimensional echocardiography.

Almost every form of congenital heart disease recognizable by ultrasound in the infant or child can be detected in fetal life. The incidence of congenital heart disease in the general population is 8 per 1,000 live births. As experience in fetal echocardiography develops, it is clearly being learned that the incidence of congenital heart disease among high-risk pregnancies rises to 75-80 per 1,000 fetuses. Even if consecutive non-high risk pregnancies are scanned, the incidence is still high, at 25 per 1,000 fetuses. Certain maternal, paternal and fetal risk factors have been identified as strong indications for performing fetal echocardiography. These include sibling or parental (either mother or father) congenital heart disease, familial or genetic syndromes, maternal diabetes and collagen vascular disease, exposure to cardiac teratogens, and some maternal infections. Fetal risk factors may appear, including abnormal appearance of the heart during an obstetric scan, extracardiac fetal anomaly, fetal arrhythmia and nonimmune hydrops fetalis. Trans-abdominal fetal echocardiography may be ideally performed from anywhere between 16-20 weeks age of gestation, with a confirmational study occasionally needed before 24 weeks. If an indication arises during the pregnancy, the procedure may be performed anytime. Transvaginal echocardiography may even be performed to as early as 12 weeks age of gestation.

The detection of cardiac abnormalities in utero has increased our knowledge of the natural history of certain pregnancies. There is a strong association between occurrence of heart defects with anomalies of other organ systems. Among pregnancies associated with cardiac defects, there is a high incidence of spontaneous intrauterine death (11.2%), high incidence of chromosomal anomalies (17%) and increased fetal loss associated with chromosomal anomalies (20%). The spectrum of congenital heart disease in the fetus is different from that seen after birth. There is a greater chance of seeing a serious cardiac disease in the former (i.e. hypoplastic left heart syndrome, atrioventricular canal, single ventricle, Ebstein anomaly), oftentimes resulting in fetal demise. Also, some abnormalities change or progress during fetal life (i.e. valvar stenosis, arterial hypoplasia, certain ventricular septal defect types, hypertrophic cardiomyopathy).

When an anomaly is detected, parents are counseled concerning the type of cardiac anomaly present and their options explained in a nondirective manner. The options are usually dependent on the gestational age at diagnosis and the presence or absence of other fetal anomalies. The prognosis of naturally carrying to term the pregnancy and the surgical options available after birth is explained. Parents are supported in their decision whatever their choice.

Knowledge of the presence of a fetal anomaly with echocardiography has been shown to allow the mother to emotionally better cope with the sick child. For the caregivers (obstetricians, perinatologists, neonatologists and pediatric cardiologists), detection of fetal cardiac anomaly allows for optimal chance of infant survival, with the expectant delivery of a high risk baby and speedy transfer to a tertiary center. Because of the high accuracy of echocardiography in detecting fetal cardiac malformations if done by experienced fetal echocardiographers, medicolegal implications may arise if parents are not allowed the chance to learn of the presence of a problem in the fetus and be given the choice of medical intervention during pregnancy. Therefore, fetuses suspected to have cardiac disease should have a fetal echocardiogram. A referral to an experienced fetal echocardiographer may allow optimal outcome. The detection of a healthy fetus or one with a heart defect would nevertheless result in proper workup and treatment.

ULTRAFAST MRI IN THE EVALUATION OF THE ABNORMAL FETUS DURING THE SECOND AND THIRD TRIMESTERS Anne M. Hubbard, The Department of Radiology and the Center for Fetal Diagnosis and Treatment, The Children's Hospital of Philadelphia With the development of ultrafast scan techniques, MRI has the potential to add valuable information to the evaluation of an abnormal pregnancy. Sequences have been developed in which an image is obtained in less than .5 sec. These include echo planar, single shot turbo spin echo, and single shot gradient echo sequences.

Prenatal MRI is useful in evaluation of the fetus with a suspected chest mass. Congenital diaphragmatic hernia (CDH) in the most common mass in the fetal chest. MRI can confirm the diagnosis by demonstrating the defect in the diaphragm and the abnormal position of the bowel. More important is the ability of MRI to determine the position of the liver, as the prognosis for survival decreases when liver is herniated into the chest. US relies on indirect signs of vessel displacement to determine the position of the liver. On MRI the liver is conspicuously seen on gradient echo images. MRI can help to diagnosis cystic adenomatiod malformation (CAM) of the lung and bronchopulmonary sequestration. The pattern of lobar involvement, the size of the lung tumor, and the amount of normal lung tissue present is more easily seen than with US. CAM and CDH may be confused on US but are easily differentiated with MRI. Lung volumes can be determined with MRI, which in the future may help determine prognosis.

Neck tumors are important to evaluate because of the potential to cause life-threatening airway obstruction at birth. MRI can differentiate teratomas and lymphangiomas, the most common neck masses, and further define the anatomy of the mass with respect to the airway and the vessels of the neck allowing advanced planning for intervention at the birth.

The fetal brain is well visualized on prenatal MRI. With US, visualization of the brain depends on the age, fetal position, and amount of amniotic fluid present. The posterior fossa is the most difficult area to see. The diagnosis of Dandy Walker malformation can easily be confirmed or excluded with MRI. Abnormalities of the corpus callosum can also be seen. Ventricular dilation is easily detected with US but the cause may be difficult to see. With MRI, the third and forth ventricles and the extra-axial spaces can be seen. Abnormalities of the brain parenchyma such as infarcts, hemorrhage, and some disorders of neuronal migration can be seen. MRI can help in prenatal diagnosis, counseling, and delivery planning.

The Current Status and Future Potential of Fetal Intervention - Image is Everything! Alan W. Flake, M.D., Director, The Center for Fetal Diagnosis and Treatment, Children�s Hospital of Philadelphia. Fetal imaging is integral to the past, present, and future of fetal intervention. In the early history of fetal intervention, the role of prenatal ultrasound was primarily to identify a fetal lesion, provide a correct anatomic diagnosis, and exclude other anatomic defects. We now depend on ultrasound, and other imaging studies, to not only provide anatomic information, but to provide prognostic, physiologic and functional information prior to surgery for optimal selection of patients for fetal intervention. Perioperative imaging allows optimal planning of the approach to the fetus and real time monitoring of fetal well being during the procedure. After fetal surgery, imaging studies allow assessment of ongoing fetal well being as well as determination of positive or negative physiologic response to fetal interventions. Prior to open fetal surgery we routinely obtain a level II ultrasound assessment augmented by power ultrasound, fetal echocardiography, and a fetal MRI. Preoperative assessment for fetuses with Congenital Diaphragmatic Hernia includes determination of the presence or absence of liver in the chest, determination of the contralateral lung area to head circumference ratio (LHR), and assessment of lung volumes by MRI to determine suitability for prenatal tracheal occlusion. Physiologic information is the primary focus for prenatal evaluation of fetuses with Congenital Cystic Adenomatoid Malformation and Sacrococcygeal Teratoma (SCT). Appropriate fetal intervention for these anomalies requires accurate early assessment of the presence or absence of hydrops and in the case of SCT, evolution of high output physiology as determined by serial Doppler blood flow measurements. The increasing use of fetoscopy for surgical interventions through small scopes with a limited field of vision requires imaging to assume an operative as well as a diagnostic role. The use of combined modalities for ultrasound guided fetoscopic interventions is being increasingly applied for procedures such as fetoscopic twin seperations, cord ligations, cystoscopic laser ablation of posterior urethral valves, and other procedures. In the future, imaging will play a progressively larger role in fetal intervention. Bronchoscopic or ultrasound guided tracheal occlusion and imaging guided tumor destruction or embolization should replace current open fetal surgery techniques. With improved resolution and technology, early gestational interventions including delivery of stem cells to the fetus and organ-directed gene therapy will undoubtedly be entirely performed under image guidance. In the past, prenatal diagnosis allowed fetal intervention to evolve. In the future, improvements in imaging and non-invasive technology will drive further expansion of indications and therapeutic options for fetal intervention.

QUANTITATIVE ASSESSMENT OF MOUSE FETAL CARDIAC ANATOMY BY MAGNETIC RESONANCE IMAGING Harvey Hensley, Susan Schachtner, Anne Hubbard, John Haselgrove & Scott Baldwin, Childrens Hospital of Philadelphia While the heart and cardiovascular systems have proven to be particularly vulnerable to genetic manipulations, current techniques for quantitative analysis of cardiovascular development are limited. Recent advances in the use of magnetic resonance imaging (MRI), with three dimensional volume acquisition, have demonstrated that this modality might prove particularly efficacious in the assessment of embryonic development. To this end, we have begun to test the feasibility of magnetic resonance microscopy in the quantitative evaluation of normal and abnormal development of the heart in mouse. Initially, embryos at 14 and 16 days post conception were injected with a contrast agent (Gd-DTPA conjugated to albumin), fixed in paraformaldehyde and then imaged at a field strength of 9.4 Tesla. T1 and T2 weighted three-dimensional data sets were acquired with a voxel size of 60x60x60 microns. The T1 weighted data sets permitted the observation of the contrast material in the vasculature and cardiac chambers, while the T2 weighted data sets allow us to distinguish the exterior of the heart from the pericardium. Registration of the data sets therefore allows us to measure ventricular wall thickness. Other quantities measured include ventricular and atrial volumes and the thickness of the ventricular septum. Image resolution and reproducibility of measurements suggests that this approach might prove valuable in evaluation of subtle as well as gross defects in cardiac development.

POWER DOPPLER IN THE ASSESSMENT OF NORMAL AND ABNORMAL FETAL VASCULARITY Beverly G. Coleman, Director, Ultrasound Imaging University of Pennsylvania Medical Center Doppler sonography is a unique, noninvasive modality for the assessment of normal and abnormal vasculature. The most widely available instruments incorporate simultaneous real-time imaging with Doppler display, which permits easy and rapid identification of vessels to be interrogated. Conventional color Doppler (CD) systems present velocity information in a color format such that flowing blood virtually serves as its own contrast agent. CD is generally based on the mean Doppler frequency shift and is a measure of directionality component of the velocity of blood moving through a volume of tissue. Changes in frequency and flow direction that occur in vascular structures in response to physiologic or pathologic states can be indirectly assessed. CD depicts areas of vascularity which are typically red or blue, although color selection is quite arbitrary. Blood flow toward the transducer is often displayed in red and flow away from the transducer in blue.

Power Doppler (PD) is a new technique in which the color map displays the strength or power inherent in blood cell motion. It has recently been described as a potential useful alternative to CD and several advantages have been reported which include the following: 1) PD does not alias because the integral of the Doppler power spectrum remains constant, which allows imaging at lower color velocity ranges thereby increasing sensitivity 3-5 times greater than CD; 2) noise on PD has a very low power signal and is displayed as soft background color; and, 3) PD is relatively angle independent. The disadvantages of PD are that it is unable to give information about velocity or flow direction and it is extremely sensitive to tissue motion. The identification of vessels with power facilitates placement of the sample volume in an area of blood flow. The identification of vessels with power facilitates placement of the sample volume in an area of blood blow. The resulting waveform obtained from pulsed Doppler can be characterized as arterial or venous and analyzed for quantitative information.

The obstetric applications of PD in the 2nd and 3rd trimester of pregnancy includes: 1) depiction of normal and abnormal fetal vascular anatomy; 2) assessment of placental pathology including previa, abruption and velamentous cord insertion; 3) evaluation of umbilical cord anomalies including single umbilical artery and nuchal cord; and, 4) investigation of flow dynamics in multiple gestations. This presentation will focus on the utility of PD in the depiction of normal and abnormal fetal vascular anatomy in structural anomalies.

Back to Table of Contents

TOWARD AN INFORMATION INFRASTRUCTURE FOR HEALTH CARE: FUNDING OPPORTUNITIES THROUGH THE ADVANCED TECHNOLOGY PROGRAM Bettijoyce Lide, Technology Administration, National Institute of Standards and Technology, Department of Commerce. The Advanced Technology Program (ATP), managed by the National Institute of Standards and Technology (NIST), works with U.S. industry to advance the nation's competitiveness and economy by helping to fund the development of high-risk but powerful new technologies that underlie a broad spectrum of potential new applications, commercial products, and services. Through cooperative agreements with individual companies or groups of companies, large and small, the ATP invests in industrial projects to develop technologies with high-payoff potential for the nation. The ATP accelerates technologies that - because they are risky - are unlikely to be developed in time to compete in rapidly changing world markets without such a partnership of industry and government. By sharing the cost of such projects, the ATP catalyzes industry to pursue promising technologies.

Within the ATP, the Information Infrastructure for Healthcare (IIH) focused program will develop critical information infrastructure technologies to enable enhanced, more fully integrated medical information systems across the healthcare industry, greatly reducing costs and errors in handling medical information. In response to the first two solicitations, 26 awards were made (out of 127 proposals submitted). These awards include over 75 participants from 24 states, and represent a commitment of $135M from the government and $142M from the private sector. Businesses of all sizes are participating, from small start-up companies to very large joint ventures. There is also heavy participation on the part of universities and non-profits. The first solicitation funded projects to pursue infrastructure development technologies; the second addressed user-interface and efficiency-enhancement technologies. A third solicitation this year yielded 94 additional proposals in these areas.

Results of this competition should be announced shortly. The original program plan calls for an additional solicitation in the area of healthcare specific technologies.

NSF SUPPORT FOR RESEARCH AND INFRASTRUCTURE IN DEVELOPMENTAL BIOLOGY Christopher Platt, Division of Integrative Biology & Neuroscience, National Science Foundation Not Available

FUNDING OPPORTUNITIES AT THE NCRR Abraham Levy, Biomedical Technology Branch, National Center for Research Resources, NIH The National Center for Research Resources (NCRR) has the primary responsibility at the National Institutes of Health (NIH) to develop critical research technologies and to provide cost- effective, multidisciplinary resources to biomedical investigators across the spectrum of research activities supported by the NIH. This infrastructure underpins biomedical research and enables advances that improve the health of our Nation=s citizens. In this presentation, the various funding mechanisms available from the Biomedical Technology area of NCRR are described. These mechanisms include: Resource Center Grant (P41); Investigator-Initiated Research Grant (R01); Innovative Approaches to Developing New Technologies (R21); First Independent Research Support and Transition (FIRST) Award (R29); Conference Grant (R13); Academic Research Enhancement Award (AREA) (R15); Small Business Innovation Research (SBIR) Grant (R43; R44); Small Business Technology Transfer Grant (STTR) (R41; R42); and Contract and Cooperative Agreement mechanisms. An overview of the varied technology developments under the current Biomedical technology program is given with special emphasis on imaging technologies. Finally, recent Biomedical Technology budgets and trends are presented.

FUNDING MECHANISMS AND OPPORTUNITIES AT THE NICHD Steven L. Klein, Ph.D. Developmental Biology, Genetics & Teratology Branch, National Institute of Child Health and Human Development, NIH The NICHD currently supports individual projects on developmental imaging via regular research grants (R01s), grants to small businesses (SBIRs and STTRs) and contracts. We are interested in expanding research in this topic by motivating interactions between individuals working in different components of the field (e.g., education, research, diagnosis and treatment, computer science, embryology, etc.). Accordingly, we will consider applications from several investigators for projects that perform research, and provide training, in multiple components of computer-assisted imaging of development. Specifically, we encourage submission of applications for scientific meetings (R13s) and recurring annual courses (T15s), Program Project Grants (P01s), Interactive Research Project Grants (IRPGs), and Institutional pre- and postdoctoral training programs (T32s). We believe that research and training activities that encompass several components will stimulate this emerging field.

Speakers and Chairs

Back to Table of Contents

Michael Ackerman, Ph.D. Assistant Director for High Performance Computing and Communications National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 301-402-4100 [email protected]	Abraham Levy, Ph.D. Biomedical Technology Branch, National Center for Research Resources, NIH [email protected]
Ernerio T. Alborilas, MD Director Non-Invasive Imaging and Fetal Cardiology Rush Children's Heart Center [email protected]	Bettijoyce Lide Advanced Technology Program, National Institute of Standards and Technology, Technology Administration, Department of Commerce [email protected]
Carmen Arbona President, MouseWorks 166 Madrone Avenue San Francisco, CA 94217 [email protected]	Elizabeth Lockett, M.F.A. Imaging Specialist, Human Developmental Anatomy Center National Museum of Health and Medicine Armed Forces Institute of Pathology [email protected]
Brian Athey, Ph.D. Director, Biomedical Imaging Programs The Environmental Research Institute of Michigan (ERIM) University of Michigan Medical School Ann Arbor, Michigan 48409-0616 313 763-6150 [email protected]	Gregory L. Merril President and CEO, HT Medical 6001Montrose Road, Suite 902 Rockville, MD 20852-4874 (301) 984-3706 x 224 [email protected]
Scott Baldwin, M.D. The Children�s Hospital of Philadelphia Department of Cardiology Philadelphia, PA 19104 215 590-1000 [email protected]	Sally A. Moody, Ph.D. George Washington University Medical Center, Programs in Neuroscience and in Genetics 202-994-2878 [email protected]
Alphonse Burdi , Ph.D. Curator, Patton Embryological Collection University of Michigan Department of Anatomy & Cell Biology Ann Arbor, Michigan 48104 313-764-4358 or 313-764-9534 [email protected]	Adrianne Noe, Ph. D. Director, National Museum of Health and Medicine, Armed Forces Institute of Pathology Washington, DC 20306-6000 [email protected]
Beverly G. Coleman, M.D, Director of Ultrasound Imaging, University of Pennsylvania Medical Center, Philadelphia, Pennsylvania. 215-662-3466 [email protected]	Christopher Platt, Ph.D. Division of Integrative Biology & Neuroscience National Science Foundation 4201 Wilson Blvd. Arlington VA 22230 [email protected]
David A. Damassa, Ph.D. Department of Anatomy and Cellular Biology Tufts University [email protected]	Tim Poe Multimedia Specialist, University of North Carolina 106 Oak Street, Carrboro, NC, 27510 919-942-5953 [email protected]
Felix DeLaCruz, Chief, Mental Retardation & Developmental Disabilities Branch, National Institute of Child Health and Human Development, NIH [email protected]	Dolores H. Pretorius, M.D. Department of Radiology, University of California, San Diego 9500 Gilman Dr., 0610 La Jolla, CA 92093-0610 619-534-1434 [email protected]
Michael D. Doyle, Ph.D. Chairman and CEO, Eolas Technologies Incorporated 312/337-8748 [email protected]	Louise E. Ramm, Ph. D. Deputy Director, National Center for Researech Resources NIH
Alan G. Fantel, Ph.D. Department of Pediatrics, School of Medicine University of Washington, Seattle, WA 98195 206-543-3373 [email protected]	Martin Ringwald, Ph.D. The Jackson Laboratory 600 Main Street Bar Harbor, ME 04609 207 288-6436 [email protected]
Alan W. Flake, M.D. Department of Surgery and Obstetrics University of Pennsylvania & Director, Children's Institute of Surgical Science, Childrens Hospital of Philadelphia [email protected]	Janet E. Rogers Mathematical and Computational Sciences Division National Institute of Standards and Technology Boulder, CO 80303-3328 (303) 497-5114 [email protected]
Raymond F.Gasser, Ph.D. Director, Computer Imaging Labs, Department of Cell Biology and Anatomy, Louisiana State University Medical Center, New Orleans, Louisiana [email protected]	Albert Shar, Ph.D. Executive Director, Computing and Educational Technology University of PA School of Medicine [email protected]
Harvey Hensley, Ph. D. Department of Cardiology, Children�s Hospital of Philadelphia Philadelphia, PA 19104 215 590-1000	Richard Simon President, Envision Development Corporation [email protected]
A. Tyl Hewitt Chief, Developmental Biology, Genetics & Teratology Branch, National Institute of Child Health and Human Development, NIH [email protected]	Bradley R. Smith, Ph.D. Department of Radiology Room 141D Bryan Research Building Research Drive Duke University Medical Center Durham, NC 27710 919-684-7852 [email protected]
Anne M. Hubbard, M.D. The Department of Radiology and the Center for Fetal Diagnosis and Treatment, The Children's Hospital of Philadelphia The University of Pennsylvania School of Medicine, Philadelphia, PA (215) 590-2560 [email protected]	David R. Soll, Ph.D. Director, WM Keck Dynamic Image Analysis Facility, Department of Biological Sciences, University of Iowa 138 Biology Build. Iowa City, IA 52242 (319) 335-1117 [email protected]
Dale S Huff, M.D. Director Of Developmental-Perinatal Pathology The University Of Pittsburgh and Magee-Womans Hospital, Pittsburg, PA 15213 412-641-1331 [email protected]	Kent L. Thornburg, Ph.D. Director, Congenital Heart Research Center Professor of Physiology & Pharmacology Oregon Health Sciences University [email protected]
Russell Jacobs, Ph.D. Biological Imaging Center, Beckman Institute 139-74, Caltech Pasadena, CA 91125 818-395-2863 [email protected]	Dan Turnbull, Ph.D. Skirball Institute, NYU Medical Center 540 First Ave, New York, NY 10016 (212) 263-7262 [email protected]
Steven L. Klein, Ph.D. Developmental Biology, Genetics & Teratology Branch, National Institute of Child Health and Human Development, NIH [email protected]	Michele Ursino HT Medical 6001 Montrose Road, Suite 902 Rockville, MD 20852-4874 (301)984-3706
Robert Ledley National Biomedical Research Foundation, Georgetown University Medical Center & Editor-in-Chief, Computerized Medical Imaging and Graphics (202) 687-2121 FAX (202) 687-1662	Betsey Williams, Ph.D. President, Muritech Inc. 100 Inman St.,Cambridge MA, 02139 [email protected]
Wesley Lee , M.D. Division of Fetal Imaging, William Beaumont Hospital, Detroit, Michigan 248-551-2071 [email protected] http://www.fetus.com/WBH and President, Fetal Imaging Resources, Inc. 248-339-9897	Elaine Young, Ph.D. Comparative Medicine Branch National Center for Research Resources NIH

Back to Table of Contents

Links to Ten Cool Projects: Human Developmental Anatomy Center, National Museum of Health and Medicine, AFIP; UC San Diego 3D Ultrasound Imaging Group; The Multi-Dimensional Human Embryo, Center for In Vivo Microsocopy, Duke University; Volume Visualization, HT Medical, Inc.; The Visible Embryo, UC San Francisco; Basic Embryology Review Program, University of PA School of Medicine; The Visible Human Project, National Library of Medicine, NIH; Internet Atlas of Mouse Embryology, MuriTech, Inc.; Gene Expression Information Resource Project, The Jackson Laboratory; Fetal Imaging Resources, Division of Fetal Imaging, William Beaumont Hospital

SOCIETY FOR DEVELOPMENTAL BIOLOGY

SDB Webmaster
Posted Tuesday, November 18, 1997