 |
Teaching Anatomy and Physiology Using Computer-Based, Stereoscopic Images
Jamie Perry, David Kuehn, and Rick Langlois
 Learning real three-dimensional (3D) anatomy for the first time can be challenging. Two-dimensional drawings and plastic models tend to over-simplify the complexity of anatomy. The approach described uses stereoscopy to create 3D images of the process of cadaver dissection and to demonstrate the underlying anatomy related to the speech mechanisms.
Virtual Reality (VR) programs (e.g., The Visible Human Project), stereoscopic, three-dimensional (3D) visual anatomy systems (e.g., Interactive Stereoscopic Virtual Reality and 3D Explorer—The Skull), and computer-generated, 3D models (e.g., Voxel Man Gallery, MPIRE, and Visible Human 3D Anatomical Structure Viewer) have been reportedly successful in biomedical instruction (Frisby 1993; Ribas, Bento, and Rodrigues 2001), simulating surgical procedures (Satava 1993; Ribas, Bento, and Rodrigues 2001; Henn et al. 2002) and diagnosis of medical conditions (Oyama 1998). Advancements in 3D technology in the medical field are important, then, as success in most disciplines of medical sciences is contingent on the mastery of complex 3D anatomy. This is also true in the area of speech science, where the focus is on anatomy related to respiration, phonation, and articulation.
Undoubtedly, learning through exploration of a human-cadaver specimen offers one of the best pedagogical experiences. It allows students to appreciate the layering of muscle architecture, and the relative sizes, forms, and shapes of structures from multiple views. However, large class sizes, increasing costs, and limited access to cadaver specimens often minimize the laboratory experience for students. Instead, some universities use animal specimens in the laboratory. While this allows students to appreciate internal anatomy, the size, shape, and orientation of the structures in lower animals are not parallel to those of the human body, and few generalizations can be made from lower animals to the human body.
When universities choose to teach without a human or animal specimen, actual observation and manipulation of the anatomy are impossible. For instance, undergraduate students learning anatomy for the first time rely primarily upon two-dimensional (2D) drawings or photographs in textbooks and atlases. Such media are valuable because they offer simple, concrete, and clearly delineated depictions of the muscle structures. However, 2D drawings can, at times, grossly underestimate or simplify complex human anatomy. Qualities such as the depth, texture, and thickness of a single muscle, and the layering of structures cannot always be appreciated. Additionally, printed drawings or photographs are not practical for showing the numerous steps involved in dissection.
Though 3D modeling can be a powerful learning tool, the major disadvantage of computer-generated models is that their production requires access, knowledge, and skill with a modeling software program—unskilled rendering artists may be unable to realistically apply surface textures such as luster, sheen, and roughness.
The introduction of VR brought another kind of 3D learning to the classroom. There are several different types of VR programs, such as interactive, 3D-rendered models; 3D-interactive models generated using magnetic resonance imaging and computerized tomography; and stereoscopic, interactive surgical-instruction tools. The major disadvantages of such programs are the relatively high costs of the software and the often technically advanced equipment, making VR systems unfeasible at some universities. In addition, the perceptual cues used to create computer-generated models do not include the depth cue that is created through binocular vision. Stereoscopic imaging, however, has been reported by many authors and credited for its relative cost efficiency and improved 3D comprehension, particularly for the instruction of anatomy (Trelease 1998; Ribas, Bento, and Rodrigues 2001; Henn et al. 2002; John 2002).
Stereoscopic imaging
Stereoscopy is the act of seeing a solid form. The interpupillary distance between the human eyes is 63 to 69 millimeters, enabling what is termed binocular vision (Judge 1926). When an object is viewed with both eyes, the gaze converges toward the object; however, the right and left eyes receive information that is specific to the right and left sides of the object, respectively. The eyes bring two slightly different representations of the object to the level of the brain, where information is combined to give the object the senses of solidity and depth. The solidity and spatial depth cues generated through binocular vision are consistent with vision in our natural world. In contrast, monocular vision occurs when the right and left eyes receive the exact same representation of the object, such as while viewing a 2D photograph.
Stereoscopic photography mimics the principles of binocular vision by creating two slightly different pictures to represent vision from both the right and left eyes individually. This can be done by using two cameras spaced slightly apart or by sliding a single camera along the horizontal plane. The distance between the two cameras or the two camera locations must be proportional to the aforementioned interpupillary distance and the distance from the viewer to the object. The amount of depth created in the image is contingent upon the amount of displacement of the camera on the horizontal plane. The most common rule used to calculate the amount of displacement is known as the one-to-thirty rule, wherein the amount of displacement is 1/30 of the distance between the viewer and the object (Frisby 1980).
Viewing stereoscopic pairs is aided by the use of special viewing devices (stereoscopes). Relatively inexpensive stereoscopes have been designed which facilitate viewing stereoscopic pairs on a computer monitor (e.g., the Lite 3D Viewer—see Resources). While looking through a viewer consisting of prism-like lenses, the eyes are able to see their corresponding picture. That is, the left eye sees the left image, the right eye sees the right image, and the brain fuses these images to provide the 3D depth perception.
While there are several cues that allow a person to determine the size, shape, and depth of an object, such as object motion, rotation of the object in space, relative size, shadows, linear viewing, and perspective viewing (Frisby 1980; Hubel 1988; Ribas, Bento, and Rodrigues 2001), stereopsis, or stereoscopic vision, is perhaps the single most important mechanism for providing depth cues (Frisby 1980; Hubel 1988).
The main disadvantage of stereoscopic imaging is that because it is based on the principle of binocular vision, people with monocular vision, or one-eye vision where two eyes are working but not in unison, are unable to see a stereoscopic pair. It is estimated that 2% of the population have monocular vision, and these individuals would be unable to experience stereoscopic, binocular viewing (Frisby 1980).
The current project uses stereoscopic imaging to provide a computer-based, 3D presentation of anatomical dissections illustrating the structures related to the respiratory, laryngeal, and articulatory speech mechanisms. The goal of the project is to enhance learning by providing students enrolled in the course Anatomy and Physiology of the Speech Mechanism at a major university with a 3D viewing experience while studying the details of human anatomy.
The design of the course creates a few notable disadvantages for students. Students attend one two-hour lab session each week, however dissections to prepare the cadaver for examination are performed by the primary lab instructor and four assistants outside of the formal laboratory session times. Therefore, students are not able to watch the process of dissection. In addition, study time during labs is limited due to the large number of students (approximately 112) per one whole cadaver. And by using one cadaver throughout the semester, students are unable to make comparisons across cadaver specimens.
This project was designed to address these limitations. The stereoscopic pairs can be viewed by students at their leisure, on their own computer screens, outside of the anatomy laboratory. Students are able to see the layering of tissue upon muscle and the complexity and interconnectivity of anatomy by watching the process of dissection through sequencing of the stereoscopic images. The 3D images provide students with the spatial depth impression that is critically important in understanding the spatial anatomy of structures under investigation.
 |
|
Figure 1. Demonstrating the
arrangement of the tripod,
arm, slide bar, and camera
mounted over the cadaver.
Additional lighting is used to
control for unwanted shadows. |
Technical details
The images are obtained from cadaver specimens used for the instruction of Anatomy and Physiology of the Speech Mechanism, a course taught to undergraduates in the Department of Speech and Hearing Science at a major university. The final stereoscopic pairs are made available to the students online before their assigned laboratory date (see Resources).
The anatomical photographs are taken using an Epson PhotoPC 3000Z 3.3 mega pixel digital camera (2048 by 1536). A Bogen friction arm (model #2929) is connected to a Bogen tripod (model #3011N) using a Bogen Super Clamp (model #2915). The camera is affixed to a Bogen Universal Sliding Plate (model #3273) with a metric scale, which is attached to the end of the arm. With the camera positioned directly over the area of interest, the first picture is taken to represent the left eye image. The camera is then moved horizontally to the right in a straight line along the slide bar to the predetermined distance set using the previously mentioned one-to-thirty rule. The typical set up is demonstrated in Figure 1.
Identification pins are placed in the specimen to help students locate the specific anatomical regions studied during each laboratory session. A metric ruler is often included in the pictures to facilitate comparisons among specimens and to provide an object size scale.
|
|
|
Figure 2. Students enrolled in
the course are able to view
the images online at their
personal computers. |
Image processing and stereoscopic programming are performed on a Dell Dimension XPS T600 computer system (600 MHz Pentium III, 512 MB RAM, and Microsoft Windows XP Professional). The two prepared images are cropped and made into one image using StereoPhoto Maker version 2.16 (see Resources).
At the beginning of the semester, students are given a stereoscope, which costs less than $2.00 (see Resources). As shown in Figure 2, images can be displayed on a personal computer. Images are intended to supplement, rather than to replace, the hands-on laboratory time and the images presented in the textbook.
Results
Feedback was solicited toward the end of the semester from students as part of an assigned, short-answer report. Students were asked to provide a written response to three questions addressing the following issues: (1) value and helpfulness of the stereoscopic project while learning anatomy and physiology, (2) suggestions to improve the project, and (3) opinion about making the images available to the general public through the internet.
At the end of the semester, the written responses provided by students were interpreted. Out of 74 students enrolled in the course, 67 completed the written, short-answer report. Sixty-four (96%) believed the project was “helpful” or “very helpful.” Several students elaborated by listing reasons for its usefulness. Three common reasons listed by students were that (1) because the images represented real anatomy, they increased students’ ability to visualize these structures, (2) the stereoscopic images were better than 2D images presented in books, and (3) the images increased the amount of time available for viewing the specimen. A few students mentioned the ability to view images from their personal home computers as an additional benefit.
Three students felt the stereoscopic images were not helpful because they could not open the picture to a size that could be viewed in 3D. This was the most common recommendation for improvement listed by students. The current website (see Resources) has been updated to decrease the amount of work required by students to format the pictures to fit the screen.
Lastly, 95% of students believe that the images should be available to the general public. The concern noted by three students about posting the images online was that it might result in improper use of the site; that is, for reasons other than scientific learning and research.
Exemplary Images
Stereoscopic imaging allows students to view the process of dissection, which until now was not feasible in this class. For example, in the area of respiration, students are required to know details of the lungs, such as the shape (size variation, number of lobes, location, and course of the fissures in the lobes), relative orientation, and function. Before students’ laboratory session, the lungs are removed and placed outside of the body for study and examination. By removing the lungs ahead of time, a comparison of the shape and orientation to other structures in the human body is difficult. However, Figures 3a, b, c, and d, which depict the removal of the lung from the thoracic cavity, demonstrate the relatively large 3D volume of the lung compared to the cavity, the concave nature of the base of the lung, and the orientation of the fissures relative to the thoracic cavity.
Figure 4 demonstrates the intricate internal anatomy of a bisected head. Students enrolled in the course are required to learn the numerous muscles and structures of this particular region during one laboratory session. Images, such as Figure 4, enable students to learn this complex 3D anatomy outside of the laboratory, increasing the amount of time they can study the actual specimen. Without viewing such images stereoscopically, it is difficult to appreciate the depth of such regions as the nasal, pharyngeal, and oral cavities.
|
Figure 4. Bisected head to demonstrate the internal anatomy. Including: nasal conchae,
hard palate, velum, pharynx, torus tubarius, salpingopharyngeal fold, spinal cord,
sinuses, and teeth. This image is to be viewed with a stereoscopic viewing device. |
|
|
Figure 5 demonstrates the use of labeling pins on the major muscles of the abdominal region. This image is particularly useful in demonstrating the layering of the external oblique abdominus over the internal oblique abdominus and transverse abdominus. In addition, students are able to appreciate the actual thickness of the muscles, which in 2D drawings tends to be exaggerated. Overall size of the muscles and location and direction of muscle fibers are clearly and accurately represented in this image.
In Figures 3–5, a stereoscope should be used for perceiving the 3D images. By holding the viewer up to the eyes and maintaining a steady focus on the images, 3D images will appear.
|
Figure 5. Demonstrating the abdominal region using identification markers to
locate the following: (1) internal oblique abdominus, (2) transverse abdominus,
(3) rectus abdominus, and (4) external oblique abdominus. This image is to be viewed
with a stereoscopic viewing device. |
|
|
Figure 6 is the same image as Figure 4, rearranged for cross-eye viewing. The left and right images switch places so that when gazed at with crossed eyes the right eye sees the proper, corresponding image on the left and visa-versa. For many people cross-eyed viewing allows stereopsis without the aide of a stereoscope. For beginners, the eye muscles may fatigue and cause some discomfort due to the unaccustomed strain. Some individuals never manage to view cross-eyed stereo images. For these reasons, it is not the preferred method of viewing and cross-eyed views are not offered to students in the class. For the purpose of this article however, the reader may enjoy attempting to view Figure 6 cross-eyed. To try this, hold the image at a comfortable distance from your face with the center in-line with your nose. Then cross your eyes. Hold the image very still and three images should appear. Without uncrossing your eyes, focus the attention of your gaze at the center image as it emerges. With a little patience your eyes and brain should adjust your perception by fusing the center image into a clearly focused, fully three dimensional view.
|
Figure 6. Same image as Figure 5, however, this version can be viewed in
cross-eyed mode. Hold the page one foot from your face and center your focus on
the space between the two images while crossing your eyes. With practice, a third
converged image will appear in three dimensions of space. |
|
|
Discussion
There is a gap between classroom learning through textbooks and real clinical practice (Henn et al. 2002), but advancements in 3D technology are a possible bridge for this gap. Henn et al. (2002) reported that a form of interactive stereoscopic viewing used for the instruction of neurosurgical procedures increases the comprehension of complex surgical procedures and speed of learning for students. The main advantage of using stereoscopic imaging over other traditional forms of technology (e.g., computer models, VR systems, and photographs or drawings) is reported consistently to be the depth cues generated from binocular vision (Judge 1926; Trelease 1998; Ribas, Bento, and Rodrigues 2001; Henn et al. 2002; John 2002). Depth cues help students understand the complexity of real anatomy and the relationship among structures, which cannot be obtained through monocular vision alone (Trelease 1998).
The current project was designed to offer three specific advantages. First, it is cost efficient. The actual creation of the stereoscopic images requires a few pieces of equipment, however, these items are easily accessible in most departments and relatively inexpensive. Second, while creation of the images does require familiarity with equipment such as a digital camera, the internet, and the StereoPhoto Maker Software, there is a minimal level of technical difficulty. Lastly, the images are located on the internet, enabling students to learn in 3D outside of the laboratory.
Overall, the stereoscopic project has been well received by students, and changes have already been implemented to address some of their aforementioned concerns. Compiled images give the viewer a 3D experience from all angles, allowing students to associate the shape and position of the facial muscles relative to the underlying internal anatomy of the maxillae and skull. The stereoscopic imaging project has been beneficial in building a photographic record of dissections and it provides students with a 3D learning system that is cost efficient, requires few technical skills to implement, and is accessible to students from their personal computers. It has notable benefits for both the instructor and students. Ideally, instruction of complex anatomy will occur in multiple modalities such as lectures, 2D drawings, hands-on laboratory time, and 3D viewing. This is important to accommodate the many different learning styles found within a single classroom.
Jamie Perry (leverich@uiuc.edu) is a doctoral candidate at the University of Illinois in Champaign. David Kuehn is a professor in the Department of Speech and Hearing Science at the University of Illinois in Champaign and Rick Langlois is a Computer Assisted Instruction Specialist in the University of Illinois Division of Instructional Technlogies and 3-D Stereo Photographer.
References
Frisby, A.G. 1993. Part I: Advances in educational technology: IVD, CD-I and journeys into virtual reality. Journal of Allied Health 22: 131–38.
Frisby, J. 1980. Seeing: Illusion, brain, and mind. Oxford: Oxford United Press.
Henn, J., M. Lemole, M. Ferreira, F. Gonzlez, M. Schornak, M. Preul, and R. Spetzler. 2002. Interactive stereoscopic virtual reality: A new tool for neurosurgical education. Journal of Neurosurgery 96 (1).
Hubel, D. 1988. Eye, brain, and vision. New York: Scientific American Library.
John, N.W. 2002. Using stereoscopy for medical virtual reality. Studies in Health Technology and Informatics 85: 215–20.
Judge, A. 1926. Stereoscopic photography: Its application to science, industry and education. London: Chapman and Hall.
Oyama, H. 1998. Virtual reality for palliative medicine. Studies in Health Technology and Informatics 58: 140–50.
Ribas, G., R. Bento, and A. Rodrigues. 2001. Anaglyphic three-dimensional stereoscopic printing: Revival of an old method for anatomical and surgical teaching and reporting. Journal of Neurosurgery 95 (6).
Satava, R.M. 1993. Virtual reality surgical simulator: The first steps. Surgical Endoscopy 7: 203–205.
Trelease R. 1998. The virtual anatomy practical: A stereoscopic 3D interactive multimedia computer examination program. Clinical Anatomy 11: 89–94.
Resources
Loreo Lite 3D Viewer—
www.loero.com.
Stereoscopic Project—
www.shs.uiuc.edu/shs300/default.htm.
Suto, M., and D. Sykes. 2002-2004. Stereo PhotoMaker. Available online at
http://stereo.jpn.org/eng/index.html.
Acknowledgments
We are grateful to the Department of Speech and Hearing Science at the University of Illinois for their support and to Mr. Mark Joseph for his technical expertise and assistance.
 Click here for PDF file.
Copyright © 2007 NSTA
|
|
E-mail to a friend
Vol. 36 No. 4, p. 18-23
