SEARCH

SEARCH BY CITATION

Keywords:

  • Chinese visible human;
  • 3D reconstruction;
  • Asian;
  • imaging;
  • computer assisted medicine;
  • CAM;
  • medical education

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. ACQUISITION OF THE CVH DATA SET
  5. 3D RECONSTRUCTION AND VISUALIZATION
  6. RESULTS AND DISCUSSION
  7. CONCLUSION AND FUTURE WORK
  8. Acknowledgements
  9. LITERATURE CITED

The United States Visible Human Project (VHP) created a digital image data set of complete human male (data acquisition finished in November 1994) and female (data acquisition finished in December 1995) cadavers in magnetic resonance imaging (MRI), computed tomography (CT), and anatomical (anatomic serial section) modes. VHP aroused worldwide enthusiasm for Visible Human Research (VHR), and the data set is being used in a variety of research and educational domains. The Visible Korean Human (VKH) male was produced in March 2001. To accelerate worldwide VHR and to promote virtual anatomy as a revolutionary break with conventional anatomy, more visible human data sets representative of different populations of the world are in demand. The Chinese Visible Human (CVH) male (created in October 2002) and female (created in February 2003) project achieved greater integrity of images, easier blood vessel identification, and were free of organic lesion (unlike the other visible human projects). We performed data acquisition, three-dimensional (3D) reconstruction, and visualization with improved technology to create CVH male and female. CVH is the first volumetric data representing a complete normal adult human male and female of an Asian population. This article presents the history of Chinese Visible Human cadavers and the methods and technology used to produce the data set. Anat Rec (Part B: New Anat) 275B:190–195, 2003. © 2003 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. ACQUISITION OF THE CVH DATA SET
  5. 3D RECONSTRUCTION AND VISUALIZATION
  6. RESULTS AND DISCUSSION
  7. CONCLUSION AND FUTURE WORK
  8. Acknowledgements
  9. LITERATURE CITED

With the coming of the digital era, digital visible human data are in demand in various fields related to human morphology, including medicine, human artificial organ research, movie and television program production, advertisement making, and the aerospace industry (Bifulco, 2002; Farre, 2002). It is difficult, however, to digitize the human body because of its intricacy. The research on digital visible human rendering, therefore, is of great significance and has many valuable applications (Sarwal and Dhawan, 2001; Temkin, 2002). In medicine, for example, a visible human can be accessed regardless of time and place, only requiring a computer and the appropriate digital medium (recorded or streaming). Organic and represented bodies converge in the visible human data sets. Digital bodies constitute one necessary medium that can be used to train future generations of students in anatomical practice and computer-assisted surgery. The visible human may also provide the basis for modern medical image interpretation and diagnosis (i.e., beta ultrasound, computed tomography [CT], magnetic resonance imaging [MRI]). The possibilities for future application of visible human data sets seem endless, e.g., radiotherapy, interventional diagnosis and interventional therapy, and development of microsurgical techniques.

As an outgrowth of the 1986 Long-Range Plan of the National Library of Medicine (NLM), the United States Visible Human Project (VHP) aims at building a digital image library of volumetric data representing a complete normal adult human male and female (Spitzer et al., 1996; Spitzer and Whitlock, 1998; Ackerman, 1999). This project serves as a cornerstone for future collections of related image data sets and as a test platform for developing methods and standards for digital-image libraries.

The creation of the U.S. Visible Human male and female signaled the starting point of virtual anatomy, which offers obvious advantages over conventional anatomy. Traditional dissection, for its educational purposes, depends almost completely on donated bodies, which are scarce, perishable, and expensive. Virtual anatomy offers “a standard model for 21st century anatomy” that enables students to obtain valuable clinical training without expensive, prepared cadavers. In addition, conventional anatomy acquires knowledge through observation and measurement after dissecting a human body, which may easily lead to lack of precision in localization of a certain organ or body structure, 3D measurement, or stereo picture. The visible human emulates anatomical illustrations, because digitization allows viewers a 3D perspective on body parts and organs. The digital cross-sections, both sagittal and coronal, can be precisely recompiled and realigned so the viewer can look through the body from virtually any plane. Furthermore, precision of measurement and localization can be guaranteed, which is the essence of modern clinical diagnosis and treatment.

There is growing worldwide recognition that Visible Human Research (VHR) provides a considerable resource for a variety of educational, medical, and private domains.

There is growing worldwide recognition that Visible Human Research (VHR) provides a considerable resource for a variety of international institutions.

As of December 1995, the U.S. VHP data set was being used by more than 1,400 research, academic, and industrial groups in 43 countries. To promote worldwide VHR with the purpose of instituting a worldwide application-oriented biomedical images library, more data sets representative of different populations of the world are in demand.

In November 2001, the 174th Xiangshan (The Fragrant Hills) Science Conference was convened for the purpose of implementing a Chinese Visible Human (CVH) project. The 42 attendees represented all medical specialties and relevant areas of computer science. They concluded that the CVH should contribute to worldwide VHR through developing a data set representing a complete normal adult male and female from an Asian population. The CVH project called for collaboration between human anatomists, experts in computer graphics and imaging, and experts in medicine.

In 2000, Korea started a 5-year Visible Korean Human (VKH) project (March 2000–February 2005).The first VKH data set was acquired in 2001 and derived from a 65-year-old patient who had died of cerebroma (Chung and Kim, 2000; Chung and Park, 2003). Therefore, it did not represent a completely normal adult human. CVH intends to augment existing VHP through its use of the following methods: First, we improved the milling machine to make its workpiece table large enough to mount a whole embedded human body, so as to avoid data loss caused by sectioning the body into blocks before cryomacrotoming. Second, to prevent small structures, including teeth, concha nasalis, and articular cartilage from falling off of the milling surface, the milling procedure was performed in a laboratory where the temperature was maintained at or below −25°C. Thus, integrity of images was ensured. A more detailed description of this procedure follows.

ACQUISITION OF THE CVH DATA SET

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. ACQUISITION OF THE CVH DATA SET
  5. 3D RECONSTRUCTION AND VISUALIZATION
  6. RESULTS AND DISCUSSION
  7. CONCLUSION AND FUTURE WORK
  8. Acknowledgements
  9. LITERATURE CITED

Our research team has been involved in the study of human sectional anatomy and computerized 3D reconstruction since 1985. We have been focusing on human thin sectional anatomy since 1990. With grants from the National Science Foundation of China for Distinguished Young Scholars, we began our research on acquiring the CVH data set in 1999. The CVH male was created in October 2002. We gained the first CVH data set using our own independently developed software and technology and were able to achieve 3D computer visualization. The international workshop on Visible Human Research was convened at the Third Military Medical University in January 2003 for an idea exchange on VHR. The data set acquisition for the CVH female was completed in February 2003, marking China's status as the third country in the world to develop a VH data set and the second country to complete a visible human male and female pair.

Cadaver Preparation

A key aspect of the CVH project was to use cadavers judged to represent as much as possible normal human anatomy. The final choices of cadavers for CVH were made from donations by the citizens of Chongqing. After the history and physical condition of the cadavers were evaluated, radiographs of the entire body were taken of those judged acceptable. If no abnormality was found in them, survey CT images were obtained of each region of the body. MRI scans were also acquired to ensure that the subjects were free of any organic diseases.

Both the CVH male and female were free of organic disease and lesion. The man whose body was used as the CVH male was 35 years old at the time of death. A citizen of Chongqing, he was 1,700 mm tall and weighed 65 kg. As for the CVH Female, the subject was also a citizen of Chongqing and was 22 years old at the time of death. She was 1,620 mm tall and weighed 54 kg. Both the CVH male and female were prepared in the same manner: To better visualize arteries and veins in 2D axial slices and allow for easier 3D reconstruction, the specimen underwent artery perfusion with 1,500 ml of 20% red gelatin solution so as to demonstrate the arterial network to the level of arterioles. Later, the specimen was placed in a box made of corrosion-resistant plates. The inner dimensions of the box were 450 mm × 500 mm × 1,800 mm. Four plastic tubes had been positioned longitudinally and straightened by the two ends, serving as fiducial rods for 3D reconstruction and as four fiducial markers for cross-sections (Figure 1). A 5% gelatin solution colored blue with food dye was used to surround the cadaver and fill this box. The box was then placed into a special freezer and frozen to −30°C for 1 week to ensure that the specimen was encased in an ice block or “cocoon.”

thumbnail image

Figure 1. Four plastic tubes (a, b, c, d) were positioned longitudinally, serving as fiducial rods for three-dimensional reconstruction and as four fiducial markers for cross-sections. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.].

Download figure to PowerPoint

Cadaver Sectioning

We used an improved TK-6350 numerical control milling machine (produced by Hanzhong Machine Tool Factory, China; designed by the engineers from our team and the factory) that allowed us to section the above-mentioned large block. Therefore, we did not have to segment the frozen cocooned (embedded in blue gelatin) cadaver into blocks and avoided data loss. This special cryomacrotome has a milling accuracy of 0.001 mm. The numerical control system was manufactured in Japan, and the mill was made in France. We milled slices of the body layer by layer, from head to toe, at −25°C in a low-temperature laboratory. The serial cross-sections were photographed with a Canon high-resolution digital camera and scanned into an animation computer. Thus, data acquisition from the cadaver was completed to obtain our structural data set of the human body (Figure 2).

thumbnail image

Figure 2. Transverse sections of Chinese Visible Human (CVH). a: Section image of head (CVH male). 1, Optic nerve; 2, mesencephalon; 3, cerebellum; 4,occipital lobe; 5, temporal lobe. b: Section image of thorax (CVH female). 1, Mammary gland; 2, right ventricle; 3, right atrium; 4, left ventricle; 5, left lung; 6, vertebral body; 7, right lung.

Download figure to PowerPoint

3D RECONSTRUCTION AND VISUALIZATION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. ACQUISITION OF THE CVH DATA SET
  5. 3D RECONSTRUCTION AND VISUALIZATION
  6. RESULTS AND DISCUSSION
  7. CONCLUSION AND FUTURE WORK
  8. Acknowledgements
  9. LITERATURE CITED

After data set acquisition of the CVH male and female, we processed 2D images in a SGI Workstation and on a P4 computer respectively. Image registration was performed through reserved fiducial rods. Reconstruction was achieved by two approaches: volume rendering reconstruction and surface rendering reconstruction. The software packages were developed by our research team.

Optimized CVH Data

According to the RGB components, the image data of the volume cross-sectional plane can be processed and displayed in color mode. By using a voxel as the basic modeling unit, we can render the body directly. We can prune, enlarge, and roam the human data in any orientation so that the inner structure of the body can be observed clearly. In this way, we can visualize the whole human body with great flexibility. To improve the rendering performance, we have optimized the data set in order that our application can be performed interactively on a single PC.

Visualization of CVH Data Set

After the visible human data were optimized, 3D reconstruction was achieved by two approaches: volume rendering reconstruction and surface rendering reconstruction based on the 2D serial sectional images that were high quality. The CVH male and female's external configuration and internal constructions can be displayed (Figure 3). With the aid of a graphics accelerator and advanced visualization techniques developed by our research team, real-time visualization of the CVH data set became feasible in a 3D virtual environment. The 3D reconstruction of Visible Human slices can be stereoscopically viewed. We can also rotate the 3D images around any spatial axis and or section them in any orientation. Based on initial transverse sectional images, we can display sagittal, coronal, and other directional sections of CVH by computerized 3D reconstruction (Figure 4) and we can also display the organs separately or as a whole. Furthermore, the distance between any two points or the angle formed by body structures can be measured accurately.

thumbnail image

Figure 3. External configuration of Chinese Visible Human (CVH) reconstructed by computer. a: Visualization of CVH male. b: Visualization of CVH female. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.].

Download figure to PowerPoint

thumbnail image

Figure 4. The coronal view of reconstructed Chinese Visible Human (CVH). a: CVH male data set. b: CVH female data set.

Download figure to PowerPoint

RESULTS AND DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. ACQUISITION OF THE CVH DATA SET
  5. 3D RECONSTRUCTION AND VISUALIZATION
  6. RESULTS AND DISCUSSION
  7. CONCLUSION AND FUTURE WORK
  8. Acknowledgements
  9. LITERATURE CITED

Acquisition of the 2,518 anatomical cross-sectional images for the CVH male took 6 months. The data set included CT, MRI, and radiographic images. Axial and coronal MRI scans of the head and neck and axial sections through the rest of the body were obtained at 3.0-mm intervals and in matrices of 512 pixels by 512 pixels. Each pixel has 256 bits of gray tone. CT data consisted of axial scans through the body at 1.0-mm intervals. CT images are 512 pixels by 512 pixels, where each 256 bits pixel value is related to electron density of the specimen at that point. The MRI and CT axial images have been aligned with the anatomical cross-sections. All slices had been sampled at a resolution of 6,291,456 (3,072 × 2,048) pixels. The data file of each section occupies 36 MB. The complete data files occupy 90.65 GB.

As for the CVH female, acquisition of the 3,640 anatomical cross-sectional images took 4 months. The visible female data set has the same characteristics as the male with the following exceptions: The serial sections had been sampled at 0.25 mm for head and 0.50 mm for other regions. CT data consisted of axial scans through the entire body at 1.0-mm intervals. The data file of each section occupies 36 MB. The complete data files occupy 131.04 GB.

We report here the first Chinese Visible Human male and female. China is now the third country with a visible human data set of its population. Because the VKH data set was derived from an old patient who had died of cerebroma, the CVH data set is also the first volumetric data representing a complete normal adult human male and female of an Asian population.

With financial support from the U.S. National Library of Medicine (NLM), a research team at the University of Colorado Health Sciences Center accomplished the world's first Visible Human Project and made the data set available for public use in November 1994. According to the U.S. VHP, the male cadaver used was dissected into four fragments, which caused data loss of the three junctions at thorax, thigh, and leg, respectively. Taking a removed left testicle into account, the visible male's body was not exactly perfect. The intervals between adjacent sections of the visible male and female are 1.0 mm and 0.33 mm, respectively. The sections were sampled at a resolution of 2,490,368 (2,048 × 1,216) pixels. The uncompressed data file of each section occupies ∼7.9 MB. The complete data files occupy 15 GB. Korea began the VKH project in March 2000, and it is expected to be finished in February 2005. The first VKH data set was obtained and made public in March 2001. The subject of the first data set was a 65-year-old male patient, and the digital photographs are at a slice thickness of 0.2 mm. The VHP and VKH data sets contribute much to VHR so that ultimately more perfect visible human data become available.

Compared with the data set of the U.S. VHP and VKH, the following features of CVH data set represent innovations: (1) Acquired from serial sections of the whole body, our data set presents no sectional data loss, whereas the data set of the U.S. VHP suffers data loss of the 3 junctions. (2) Our digital image achieves a higher resolution than that of the VHP. (3) Vascular perfusion was performed to allow easier blood vessel identification. (4) The milling procedure was performed in a low-temperature laboratory to prevent small structures (including teeth, concha nasalis, and articular cartilage) from falling off of the milling surface, to ensure better image integrity. (5) The subjects selected were of moderate age, size, and height so as to be more representative of the population. The CVH male and female reported here take a step forward in using advanced equipment and are representatives of a more complete and accurate visible human data set.

CONCLUSION AND FUTURE WORK

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. ACQUISITION OF THE CVH DATA SET
  5. 3D RECONSTRUCTION AND VISUALIZATION
  6. RESULTS AND DISCUSSION
  7. CONCLUSION AND FUTURE WORK
  8. Acknowledgements
  9. LITERATURE CITED

Encouraged by the theoretical importance and vast number of applications of VHR, Chinese scientists in both computer science and informatics have used the data set made public by the U.S. VHP to perform fundamental research on developing a Chinese Visible Human data set; including data compression, image segmentation, image registration and fusion, geometric modeling, 3D reconstruction, and drawing. Scientists in informatics have collaborated with scientists in medicine to carry out fruitful research in various fields, such as 3D display of images, virtual endoscopy and surgery, surgical navigation, frameless stereotactic brain surgery, virtual environment for acupuncture learning and practice of traditional Chinese medicine, and creating a virtual surgical environment with tactile feedback. Some of these advances have made their way into clinical practice.

The collaboration between Chinese anatomists and computer scientists has allowed us to reach the stage of research on 3D reconstruction and visualization of multiple loci, organs, tissues, and cells.

The collaboration between Chinese anatomists and computer scientists has allowed us to reach the stage of research on 3D reconstruction and visualization of multiple loci, organs, tissues, and cells.

The previous work laid the basis for research on CVH. The successful acquisition of the first CVH data set marks an anthropological extension of the world's existing visible human database and accelerates the development of methods and standards for digital-image libraries.

It is our hope that the Chinese Visible Human project we are developing will make innovative advancements in the better demonstration of microvessels and peripheral nerves. We have signed a collaboration agreement with scientists from the U.S. and Korea to cooperate and make additional contributions in VHR. We would be very interested in establishing further regional and international collaborations to foster research and development in this important area.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. ACQUISITION OF THE CVH DATA SET
  5. 3D RECONSTRUCTION AND VISUALIZATION
  6. RESULTS AND DISCUSSION
  7. CONCLUSION AND FUTURE WORK
  8. Acknowledgements
  9. LITERATURE CITED

The authors thank investigators at Virtual Reality, Visualization and Imaging Research Center, Chinese University of Hong Kong for data set visualization. We thank computer experts at the Department of Computer Science and Technology, Tsinghua University for their support. We also thank our domestic colleagues for their dedication to Chinese Visible Human research that made this project successful. We especially acknowledge those citizens who donated their remains to medical research.

LITERATURE CITED

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. ACQUISITION OF THE CVH DATA SET
  5. 3D RECONSTRUCTION AND VISUALIZATION
  6. RESULTS AND DISCUSSION
  7. CONCLUSION AND FUTURE WORK
  8. Acknowledgements
  9. LITERATURE CITED