Teaching medical histology at the University of South Carolina School of Medicine: Transition to virtual slides and virtual microscopes

Authors


  • The authors are reproductive biologists and members of the Department of Cell and Developmental Biology and Anatomy at the University of South Carolina School of Medicine, where they currently teach medical histology. Dr. Blake, Professor, is currently President-Elect of the Society for Experimental Biology and Medicine and Secretary-Treasurer of the Association of Anatomy, Cell Biology and Neurobiology Chairpersons. He was given the Teacher of the Year Award by the first-year medical students in 2000, 2001, and 2002. His research interests are in reproductive neuroendocrinology. Dr. LaVoie, Assistant Professor, also was recognized for her excellence in teaching with the prestigious String of Pearls Award from the first-year medical students. Her research interests focus on ovarian follicular development and function. Dr. Millette, Professor, is the Course Director of Medical Microanatomy and former Head of the Curriculum Committee of the School of Medicine. His research interests are in cell–cell interactions within the testes.

Abstract

We describe how the histology course we teach to first-year medical students changed successfully from using glass slides and microscopes to using virtual slides and virtual microscopes. In 1988, we taught a classic medical histology course. Subsequently, students were loaned static labeled images on projection slides to introduce them to their microscope glass slides, and we made laser disks of histological images available in the teaching lab. In 2000, we placed the static labeled images and laboratory manual on the Web. We abandoned the Web-based approach in 2001. Faculty selected specific areas on microscope glass slides in student collections for scanning at a total magnification of 40, 100, 200, or 400. Christopher M. Prince of Petro Image, LLC, scanned the glass slides; digitized, encoded, and compressed (95%) the images; and placed them on CD-ROMs. The scanned images were viewed up to a magnification of 400 using the MrSID viewer (LizardTech software) and the computer as a virtual microscope. This viewer has many useful features, including effective microscope and telescope functions that provide greater versatility for sample study and speed in localizing structures than was possible with the actual microscope. Image detail is indistinguishable from that viewed under the light microscope at equivalent magnifications. Static labeled images were also placed on CD-ROMs to introduce students to the virtual slides. Students could view all the images on their CD-ROMs at any time and in any place with their laptop computers without going online. Students no longer rented light microscopes in 2002. Both students and faculty have shown strong support for using this approach to teaching histology during the past 2 years. Anat Rec (Part B: New Anat) 275B:196–206, 2003. © 2003 Wiley-Liss, Inc.

AN OVERVIEW OF TEACHING MEDICAL HISTOLOGY IN THE 20TH CENTURY

Medical histology has been a long-standing basic science course in the medical school curriculum worldwide. Changes in histology course materials during the 20th century have reflected improvements in histological techniques and slide preparation as well as developments in light microscopes and associated photomicroscopy.

Changes in histology course materials during the 20th century have reflected improvements in histological techniques and slide preparation as well as developments in light microscopes and associated photomicroscopy.

Transmission and scanning electron photomicrographs were used in teaching histology during the second half of the 20th century. Changes in course content during the 20th century initially emphasized new knowledge of structure as observed at the light and electron microscope levels. Faculty members subsequently incorporated more histophysiology and histopathology into their courses to emphasize newly acquired information on the function and clinical relevance of the cells and tissues being studied. The presentation of a significant amount of cell biology also has been incorporated into textbooks and courses. Changes that were incorporated during the 1980s and 1990s have occurred at the same time as an emergence of pressures from the Liaison Committee on Medical Education (LCME) and local university administrators to decompress the curriculum and reduce student–faculty contact hours in courses, including histology. At a significant number of medical schools, financial constraints have resulted recently in only partial replacement of retiring faculty, and the teaching loads of remaining faculty, therefore, have increased.

During the latter part of the 20th century and the beginning of the 21st century, there has been a rapid acceleration in the use of computers and the Web in medical school teaching, including the teaching of histology. Technology has been developed to scan histological specimens and to digitize, encode, and compress the images. The images then can be stored on the Web or disks and subsequently retrieved and studied with computers using readily available viewers. In this article, we report how the histology course we teach to students during their first semester in medical school at the University of South Carolina has changed from using microscope glass slides and light microscopes to using virtual slides and virtual microscopes. These changes in imaging have been successful in addressing several concerns of our students, faculty members, and administrators to meet challenges of training our future physicians in the 21st century.

CHANGES IN TEACHING MEDICAL HISTOLOGY AT THE UNIVERSITY OF SOUTH CAROLINA (1977–1999)

From the fall of 1977, the time when the first class of medical students matriculated at the University of South Carolina, until 1988, faculty members at the University of South Carolina were teaching what was considered throughout the country to be a classic histology course (Hightower et al., 1999). Students purchased a textbook, and faculty presented lectures that used projection slides, chalk and the blackboard, and overhead projections. Most faculty members provided the students with additional handouts. Students also purchased an atlas and were given a laboratory manual written by the faculty. In addition, they were loaned a collection of glass slides that they viewed with light microscopes that they rented. The only substantive change incorporated into the course at the University of South Carolina between 1977 and 1988 was in the required textbook. In keeping with a reduction in course hours allotted to medical histology during this period, a switch was made to a text that contained less detail than the tome that had been used previously. Faculty members who had been teaching the same material for many years resisted changing the textbook. However, this change was important, timely, and necessary at the University of South Carolina to have realistic expectations of what students could learn in a limited period of time and needed to learn for their future roles as physicians.

One of us (C.A.B.) previously taught medical histology at the University of Nebraska Medical Center. At Nebraska, faculty members had used labeled 35-mm projection slides most effectively to introduce students to microscope glass slides. These were made originally to construct self-study modules (Bauer et al., 1976). In 1989, faculty developed and incorporated 35-mm projection slides into our course at the University of South Carolina. Most of the projection slides were photomicrographs of tissues in the students' microscope glass slide collections. Tissues and structures on the projection slides were labeled. We then loaned students the projection slide collections along with mini-projectors and mini-screens and provided them with a key to the labeled structures in their laboratory manuals. These projection slides served to introduce students to the glass slides in a manner that was identical for all students. The projection images were well received by the students and their incorporation into the course significantly reduced the need for faculty to answer questions during scheduled laboratory hours.

Since 1992, students have been required to conduct limited library research and write a 3-page, take-home essay that required critical evaluation of histologic images (Hightower et al., 1996). Essay topics constructed by faculty often focus on a current newsworthy medical condition, include if possible a medical dilemma, and always emphasize a cytological or histological component. Students are provided a choice of questions and are encouraged to choose a problem consistent with their intended specialties or with their life experiences. This exercise has linked histological images with clinical issues and has been received well by the medical students.

In 1995, the department purchased Histology: A Photographic Atlas by Stephen W. Downing, a bar code reader, and a laser disk player. The atlas is a laser disk of labeled histological images developed for use in a multimedia-based laboratory at the University of Minnesota School of Medicine in Duluth (Downing, 1995). These materials were made available for optional review for individual or small group viewing in a separate section of our teaching laboratory. Some students routinely used the laser disk with its excellent images, primarily outside of scheduled laboratory periods and in preparation for laboratory practical exams. The positive response of students to this technology provided an early impetus for faculty development of more technologically advanced methods for the transferal of histological data.

A TECHNOLOGY-ENHANCED CURRICULUM AND UTILIZATION OF THE WEB IN TEACHING MEDICAL HISTOLOGY (2000)

There were approximately 70 entering medical students in the years 2000–2002 at the University of South Carolina. During those years, the medical histology course we taught was scheduled for 50 h of lectures, 53 h of laboratory, 4 h of review lectures, 2 h for 3 quizzes, and 12 h for 4 written and laboratory practical exams that included a comprehensive final.

In 2000, the University of South Carolina School of Medicine instituted an initiative termed the Technology-Enhanced Curriculum. As a result of this initiative, entering medical students were required to purchase laptop computers with specific hardware and software requirements. In addition, a wireless network was installed that covered the campus. Thus, students were able to access the Internet at any time and from any location on campus. Most faculty and students have embraced this technology-enhanced curriculum.

That year, faculty members teaching histology made the following changes. They uniformly prepared lectures as Microsoft PowerPoint presentations, delivered them in the lecture hall, and also placed them on the medical school's Web site where they could be downloaded, annotated, and printed by the students. In addition, they placed the static labeled images that were on the projection slides on the Web site in Web page format. Students no longer received loans of the projection slide sets, mini-projectors, and mini-screens. Electron photomicrographs were scanned and also placed on the Web. The laboratory manual was placed on the Web. These changes resulted in students being able to view lecture and laboratory images at any time through Internet access. The changes also resulted in the department not having to spend time and money in maintaining the projection slide sets, mini-projectors, and mini-screens.

Overall, student reaction to the use of the Web was positive. However, difficulties in accessing the Web or speed of downloading images were often mentioned. Frequent complaints centered on the server being down or the long time associated with downloading images from home using dial-up. In addition, the majority of the students made the decision to print the laboratory manual from the Web.

THE MOVE TO VIRTUAL SLIDES AND VIRTUAL MICROSCOPES (2001)

Within 1 year, we stopped using the Web for teaching except to continue to place our PowerPoint presentations there. The reason was the very rapid move we made during a 3-month period in the spring and summer of 2001 to prepare virtual slides for use with a virtual microscope for the incoming class in August of 2001. The virtual slides were stored on CD-ROMs. The laboratory manual was no longer placed on the Web. Students received a printed version. Static labeled images at the light and electron microscope levels were placed on CD-ROMs as individual Web pages and were no longer placed on the Web. Instead of being used to introduce students to the microscope glass slides, they functioned to introduce students to virtual slides.

To prepare images for virtual slides, we first selected specific areas on the microscope glass slides in the student collections for scanning. These tissue areas were digitized by using a proprietary method originally developed for the quantitative and spatial analysis of geological specimens in the petroleum industry by Christopher M. Prince, Ph.D., of Petro Image, LLC (www.petroimage.com) in Columbia, SC. His method differs from systems that rely on precise stage positioning to acquire nonoverlapping images (Romer et al., 2003). He uses an advanced tiling algorithm whereby up to hundreds of individual scanned fields are merged into a final, single, seamless, composite. As the digitizing program does not assume any particular fixed overlap between individual scanned fields, it is hardware independent. He used a standard motorized X,Y scanning stage implemented on a research grade Olympus light microscope equipped with 4, 10, 20, and 40× objective lenses. Total magnification was 40, 100, 200, and 400×, respectively. In some cases, the entire tissue section on a microscope glass slide was selected for scanning. Potential image resolution is limited only by the optical system. Scan times are short, being measured in minutes. Currently, Petro Image LLC scans slide sets on a commercial basis and sells the hardware and software needed for individuals to scan their own slides.

After slides were scanned and final composites were merged, resultant image files were encoded and compressed by using freely available programs developed by a company called LizardTech. The programs are known as MrSID. Encoding into MrSID format is straightforward and rapid, requiring only seconds for most files. Encoding results in a 95% reduction in file size with no significant loss of image quality, color balance, or resolution. The compressed files were placed on CD-ROMs. Images on CD-ROMs are then decompressed and viewed on computers using the stand-alone MrSID Viewer. The MrSID programs facilitate transfer of information to students using their computers and the CD-ROMs containing image files relevant to the prepared laboratory exercises.

Images may be opened on virtually any computer or appropriately connected monitor. Both PCs and Apple Macintosh computers may be used, although we used the PC format. In addition, other MrSID programs, including a plug-in for Adobe Photoshop are available. As a result, digitized and encoded images may be annotated and altered as desired for use in examinations, review packages, or research applications.

A total of 377 static labeled images and 326 virtual slides were placed on CD-ROMs. A static image and a virtual slide used approximately 4 kB (3–12 kB) and 4 MB (0.5–13 MB) of space, respectively. A total of four CD-ROMs were produced to correspond to the four sections of our medical histology course. Two CD-ROMs (approximately 650 MB capacity each) would have been sufficient to store the images. However, the CD-ROMs were produced and given to students sequentially because of time restraints in completing them for the first incoming class that would use them. Image file size and number of images are becoming not a matter of concern as DVD-ROM drives have entered the marketplace at affordable prices. DVD disks, which routinely hold up to 4.7 GB of data, have markedly more capacity than do CDs.

Students were loaned the CD-ROMs containing both static labeled images and virtual slides. The students could view the static labeled images on their computers by using the Web browser to open the Web pages. The virtual slides were viewed up to 400 total magnification by using LizardTech software and the laptop computer as a virtual microscope. Both static and virtual images could now be viewed at any time and in any place with their computers without going online.

Anticipating less need for light microscopes, we recommended that pairs of students rather than individual students rent a light microscope during the fall of 2001. Students were not discouraged from using their light microscopes for evaluating all glass slides. However, we only required that they do so for viewing some cells and structures at >400 power. In this way, we ensured that students learned to use light microscopes. We did not recommend students rent microscopes in 2002. The reasons for this approach were threefold. First, microscope glass slides and light microscopes were little used in 2001. Students preferred to use the virtual slides and virtual microscopes. Second, we made a small number of microscopes that we had in the department available to the students. These proved sufficient for the 5 h of scheduled laboratory study of the connective tissue cells, bone marrow, and peripheral blood that required use of the 100 power objective with oil immersion. The third reason that we did not recommend students rent microscopes was because our supplier, a company located in Charleston, South Carolina, was no longer in that business. This finding may represent a trend in the United States as a colleague in Chicago communicated that they had been using a different supplier who also was no longer in business.

VIEWING STATIC LABELED IMAGES

Static labeled images can be viewed rapidly after insertion of the CD-ROM into the computer. Web pages appear almost instantaneously after double-clicking on labeled icons. For student convenience, figure legends are included with the static images. The legends are also printed in their laboratory manuals along with more detailed information.

Presently, we are replacing original static labeled images made from photomicrographs with static labeled images made from the virtual slides. Specific areas on the virtual slides are captured by using the MrSID plug-in program for Adobe Photoshop. The encoded images are then annotated and saved as Web pages. The scanned images are comparable to those originally obtained by photomicroscopy. Students are expected to have less difficulty in using the labeled examples on images from scanned tissue than those from photographed tissue. The labeled images from scanned tissue have been made from the original virtual slide that was replicated for all students. In contrast, the labeled images of tissue obtained by photomicroscopy were made from microscope glass slides that were not used to make the virtual slides.

An example of a static labeled low-power image of mammalian jejunum made from a virtual slide is shown in Figure 1. When deemed appropriate, we include static labeled images at higher powers as shown in Figure 2.

Figure 1.

This image shows a cross-section of a partial mammalian jejunum originally scanned at 400×. The lumen is at the top of the field. The arrows designate major layers of the gut wall: mucosa, which is composed of the epithelium, lamina propria, and the muscularis mucosa (a); submucosa (b); muscularis externa, consisting of the inner circular and outer longitudinal layers of smooth muscle (c); and serosa (d).

Figure 2.

This image is a higher power magnification of the same jejunum originally scanned at 400× and shown in Figure 1. This image field shows transition between the mucosa (m) and the submucosa (sm). The muscularis mucosa (mm) is seen in the center of the field. Arrows indicate the position of Paneth cells at the base of the crypts of Lieberkuhn.

MrSID Viewer

The MrSID Viewer is freely available at the LizardTech Website (www.lizardtech.com). It has many important and user-friendly features that permit the computer to pretend it is a microscope. One can familiarize themselves with its capabilities by looking carefully at the MrSID viewing box (Figure 3).

Figure 3.

This image is the MrSID viewing box. The toolbar in the upper portion of the figure has been copied, enlarged, and pasted lower inside the viewing box for better visualization. Moving from left to right, the top menu buttons include icons for opening files, exporting images as TIFF files, printing, quick access to the full image data set, real-time panning, standard zoom-in/zoom-out function, point-to-point distance measurement, a microscope function, a telescope function, a clipboard, and online help files. At the bottom, cursor position coordinates and image magnification scale information are visible at all times. Standard scroll bars are provided in both X and Y coordinates. Finally, on the far right, a continuously adjustable zoom slider provides zooming to any intermediate level. One is not restricted to the designated zoom points used by the top menu buttons. This figure also shows the computer screen as it first appears after opening a digitized file encoded in the MrSID format. This sample is a section of human jejunum originally scanned by using a 40× objective lens and a 10× ocular. It cannot be readily identified at this magnification.

Viewing Virtual Slides with the Virtual Microscope

The computer screen as it first appears after opening a digitized and encoded image file is misleadingly most unimpressive and is shown in Figure 3. This sample is a section of jejunum originally scanned by using a 40× objective lens and a 10× ocular and is the same plastic-embedded tissue that was used to prepare the static labeled images shown in Figures 1 and 2.

By using the standard zoom-in or continuously adjustable zoom slider feature, the same image file can be enlarged enough so the entire data set is visible on screen (Figure 4). The image is comparable in size to that which would be observed by using a 4× microscope objective lens. The image now can be readily identifiable as jejunum. With yet additional zooming in to attain approximately one quarter of the final possible magnification for this image one can see increased detail of the intestinal crypts (Figure 5). When zooming in on fields of view that are less than the full data set, it is advantageous to activate the telescope in the top menu to open a window inset (Figure 5). This inset shows a box outlined in red delineating the exact position of the area being viewed on screen with respect to the entire viewable field. This is a most valuable feature for orienting one's field of view.

Figure 4.

This is the same image file of jejunum as that shown in Figure 3. It has been enlarged enough so that the entire data set is visible on screen, and the image is comparable in size to that which would be observed by using a 4× microscope objective lens.

Figure 5.

This is the same image file of jejunum as that shown in Figures 3 and 4. It is approximately one quarter of the final possible magnification for this image. The telescope in the top menu was activated and a window inset is open. The inset shows a box outlined in red delineating the exact position of the area being viewed on screen with respect to the entire viewable field.

One has complete control over the placement of the telescope box by grabbing the interior or perimeter of the inset box with the cursor and dragging. One also can alter the size and position of the box by clicking on a desired area within the telescope box and dragging to create a box of interest. The box has been reduced in size, and its position altered in Figure 6. In real-time, the whole screen image automatically and rapidly adjusts to conform. Figure 6 shows the image adjusted to half of the final possible magnification for this image. This telescope capability proves superior to standard panning techniques used when viewing actual glass slides.

Figure 6.

This is the same image file of jejunum as that shown in Figures 3–5. After accessing the telescope box shown in Figure 5 with the cursor and clicking and then dragging it to box in an area of interest, the size of the box has been reduced and its location altered. The whole screen image automatically adjusted to conform. It adjusted to half of the final possible magnification for this image.

By using the zoom-in function or the adjustable zoom slider, one can see the maximum resolution of this image file (Figure 7). Image detail is indistinguishable from that viewed under the microscope at 400×.

Figure 7.

This is the same image file of jejunum as that shown in Figures 3–6. By using the zoom-in function or the adjustable zoom slider, one can see the maximum resolution of this image file. Image detail is indistinguishable from that viewed under the microscope at 400×.

The MrSID microscope function can be used when the image is not displayed at maximum resolution. In Figure 8, the image has been reduced to one quarter maximum resolution. Activating the top menu microscope button opens a window inset that displays at maximum zoom factor the area immediately surrounding the point of the arrow cursor. The size of the inset box is user controlled by simply clicking and dragging the inset box sides or corners. The microscope function represents one clear advantage of computer viewing compared with traditional microscopy. The function is very helpful for faculty when interacting with students one-on-one in the laboratory.

Figure 8.

This is the same image file of jejunum as that shown in Figures 3–7. It is approximately one quarter of the final possible magnification for this image. Activating the top menu microscope button opened a window inset that displays at maximum zoom factor the area immediately surrounding the tip of the arrow cursor (see cursor on larger image). The size of the inset box, that is the extent of area surrounding the cursor to be viewed, is user controlled by simply clicking and dragging the inset box sides or corners.

By using the pen in the upper toolbar, one can measure point-to-point distances on the image. However, to date, the MrSID Viewer does not include biologically relevant units, such as micrometers, in the software.

EXAMINATIONS

We made a major change in the way we examined students for their laboratory practical exams concomitant with the debut of virtual slides and microscopes in the course. We no longer used light microscopes or proctored the practical exam in the laboratory. Rather, some faculty used the teaching laboratory for half of the class to take a written exam with multiple-choice questions, whereas other faculty simultaneously projected images in the lecture hall for the practical exam. The majority of questions on the practical examinations used the virtual slides in the students' collection. This encouraged students to study their virtual slides, and all students had identical slides. Faculty, therefore, avoided previous student complaints that the particular slides in their individual collection were not as good as those of a fellow student, as occasionally occurred when the slide collections consisted of glass slides. Faculty members appreciated using virtual slides for examination. The slides provided numerous images to choose from, and the images could be moved and shown at different powers during examination. Moreover, images could be prelabeled in Adobe Photoshop or PowerPoint and used in examinations as static images of the virtual slides in their collections. Thus, students could not complain that they were tested on images that they were not previously asked to study.

STUDENT EVALUATIONS

Students were asked to volunteer evaluations of the use of the maneuverable MrSID images after the end of the course in 2001. They were asked not to volunteer unless they agreed to fill out the questionnaire conscientiously. Thirty-three of 70 students responded. Shortly before the end of the course in 2002, students were asked to fill out a similar questionnaire. This occurred at the time they were required to fill out course evaluation forms at the beginning of a scheduled review in preparation for the final exam in the course. Fifty-seven of 70 students showed up for the review and filled out the questionnaire. Students did not sign the questionnaires, and anonymity was maintained both years.

Students were given a series of statements on the questionnaire and asked to mark whether they agreed strongly, agreed, were neutral, disagreed, or disagreed strongly. Students responded to the questionnaires similarly in both years. The statements are shown at the top of the histograms in Figures 9–11. The first four statements (Figure 9a–d) were given to students only in 2001. They were related to the use of microscopes, and the students in 2002 did not rent microscopes. The last two statements (Figure 11c,d) were new and given only to students in 2002.

Figure 9.

a–f: Student responses to statements in a questionnaire are plotted. Students in the 2001 class only were given four of the statements (a–d). Students in both the 2001 and 2002 classes responded to the two additional statements shown in e,f. Note that the student number responding for the class of 2001 was 33 and that for 2002 was 57. See text for details.

Figure 10.

a–f: Student responses in 2001 and 2002 to statements in a questionnaire are plotted. Note that the student number responding for the class of 2001 was 33 and that for 2002 was 57. See text for details.

Figure 11.

a–d: Student responses to statements in a questionnaire are plotted. Note changes in the X-axis. Students in the 2002 class only were given two of the statements (c,d). Note that the student number responding for the class of 2001 was 33 and that for 2002 was 57. See text for details.

The first statement was “I found it necessary or desirable to use BOTH the light microscope and the laptop computer often during the semester.” Only approximately 25% of the students found it necessary or helpful to use both the microscope and the computer (Figure 9a). Over two thirds of the students found navigation with the computer and MrSID viewer to be easier than that with the light microscope and glass slides (Figure 9b). Students agreed that the laptop approach saved time compared with using the light microscope (Figure 9c). In general, students advocated a decrease in the number of microscopes to be shared (Figure 9d). In fact, two thirds of the students advocated the elimination of the light microscope (Figure 9e). Only eight of the students in 2001 indicated that we should not get rid of the microscopes completely. The students in 2002 were asked this question as well even though they only used the departmental light microscopes for a limited number of labs. Of interest, 87% of the 57 students who responded in 2002 advocated getting rid of light microscopes completely (Figure 9e). Almost all students found the MrSID viewing technology effective (Figure 9f).

Of importance, 88% of the students in 2001 and 95% of the students in 2002 found the maneuverable images viewed with the MrSID viewer to be of sufficient resolution to identify structures specified in their assignments (Figure 10a). In both years, there was virtually no difficulty in interfacing the printed instructions with the computer images (Figure 10b). Only one student in 2001 and two students in 2002 disagreed. Students also were virtually unanimous in advocating the ability to conduct the lab exercises on their own time (Figure 10c). The same three students that disagreed with the last statement disagreed again. Students also strongly preferred the CD-ROM to a Web-based approach to view images (Figure 10d). There was considerably more enthusiasm for using their laptops outside of class than during scheduled laboratory class (Figure 10e cf. Figure 10f).

Please note that the X-axis has been changed for Figure 11a–d. Overall, students believed that the use of laptops increased their grades (Figure 11a). Most students spent 3–5 h weekly using their laptops to conduct laboratory exercises in histology (Figure 11b). On the average, 5 h of laboratory with faculty present were scheduled weekly. In the laboratory, students worked primarily in groups of two or three, although a few worked in groups of more than three (Figure 11c). Outside of the laboratory, students primarily used the laptops to view images alone, although a significant number studied with one other (Figure 11d).

DISCUSSION AND CONCLUSIONS

The imaging and presentation protocols employed by our students make excellent use of the computer and allow integration of classic and technology-based methods in histology.

The imaging and presentation protocols employed by our students make excellent use of the computer and allow integration of classic and technology-based methods in histology.

They also are applicable to other courses and disciplines that rely heavily on images. For example, the techniques are readily expandable to courses in neuroanatomy or pathology, establishment of archives of images of tissues on slides prepared for clinical evaluation of possible pathology, and the study and archiving of light microscope images used in research.

These technological changes occurred, in part, as a response to extra-departmental pressures that affect the teaching of all medical school courses, including histology. The Association of American Medical Colleges' 1984 publication Physicians for the Twenty-First Century: The GPEP Report emphasized active learning formats and instilling principles of self-learning and problem-solving capabilities in students. The LCME, the nationally recognized accrediting authority for medical education programs leading to the M.D. degree in U.S. and Canadian medical schools, and local institutional administrators recommended or required decompression of the medical curriculum with consequent reductions in student contact hours with faculty. In addition, the utility and value of light microscope analysis of glass slides, as traditionally conducted in the teaching laboratory, are often questioned in light of modern technologies. Increased teaching loads also have stimulated faculty to develop new strategies that would reduce faculty contact hours without compromising the educational experience of the students.

The use of static labeled images to introduce students to the virtual slides facilitates them spending more time self-learning and less time in scheduled laboratories when faculty members are present. These static images are well received by students, faculty, and fourth-year medical student teaching assistants. In particular, they bolster uniformity in emphasizing to the students precisely what they are expected to learn. The faculty to student ratio can be reduced during laboratory hours without compromising the quality of student education.

Overall, students rated the computer images as viewed with MrSID as having excellent resolution. As a group, they had little enthusiasm for the light microscope. They found it easy to navigate and interface the printed lab handouts with the images. They were very positive about having the images on CD-ROMs that, unlike the Web, they can access at any time. Moreover, students believed that, compared with the use of microscopes, the computer images saved them time and enhanced grade performance.

A study was conducted at the University of Iowa in the spring of 2000 to assess the potential effectiveness of using virtual slides on the Web in their medical histology course (Harris et al., 2001). A total of 19 virtual slides were made from microscope glass slides in the endocrine, urinary tract, and male genital tract units in student collections. These virtual slides and a viewer were made available to students on the Web in addition to the microscope glass slides from which they were made. Students compared the two viewing approaches and rated the virtual slides and virtual microscope highly. Overall, they preferred the virtual microscope laboratory. Subsequently, all slides in the student collections (nearly 130) were made into virtual slides and made available to the students on the Web as an option for learning the histological material. Students again rated the virtual slides and microscopes very highly (Heidger et al., 2002).

From the student point of view, the advantage of using CD-ROMs compared with the Web to store images is the ability to view them at any time and in any place with a laptop computer without going online. From the faculty point of view, the disadvantage of using CD-ROMs compared with the Web is the inability to readily add new or altered images. It seems likely that judicious application of both Internet-based and CD-ROM–based approaches will prove beneficial. One should also not discount the potential future advantages of DVD recording techniques in terms of storage capacity, video integration, and similar modalities.

Our current procedures using virtual slides and virtual microscopes provide important advantages. Image quality is not significantly compromised if proper optical systems, commonly available, are used. Computer-aided image observations retain all features of light microscopic evaluation. Telescope and microscope functions provide even greater versatility for sample study than does the actual microscope and localization of structures on virtual slides with the virtual microscope is more rapid than localization of structures on glass slides with the light microscope. Moreover, images may be standardized and archived for instruction, performance evaluations, or research. Space is no longer needed to store light microscopes and glass slide collections. Because students are required to purchase laptop computers, it is cost-effective to incur the initial costs of producing virtual slides compared with having to maintain glass slide collections and rent or maintain light microscopes. The technology also provides the ability to increase the number of virtual slides in any individual institution's collection by sharing digital images with others. This avenue is currently being actively explored by the University of Iowa and by a joint initiative between the University of South Carolina School of Medicine in Columbia and the Medical University of South Carolina in Charleston. This is not possible with glass slide collections.

By using our approach, students are encouraged to develop the life-long learning skills deemed so important in current educational practice. Note also that students access their CD-ROMs whenever and wherever they chose—even while eating at the local pizza parlor or flying home during breaks. We also discovered that this approach allowed a decrease in the faculty to student ratio in the laboratory without compromising the educational experience of the students. This finding can be attributed to students teaching each other as they huddle around computers, faculty answering questions for groups of students rather than individuals, and methodology that facilitates students learning on their own. Importantly, our students clearly liked using this approach to teaching histology, as do the faculty.

We do believe that it is important that students gain some experience in learning to use a light microscope during their medical education whether it is in histology or in some other course or venue. However, we now are in a position to eliminate completely the light microscope from teaching in histology. The technology used by Christopher Prince of Petro Image works superbly for scanning at 1,000× and in different focal planes. He also has developed software that allows one to focus with the virtual microscope. In addition to being developed originally for the petrochemical industry, his protocols have been used by geologists to analyze sections of deep glacial ice cores and by the electronics industry to assess the integrity of silicon wafer microchips. Our application, now used successfully for 2 years in teaching histology, is the first directed toward biological specimens.

Medical histology has been taught very successfully without microscope glass slides and light microscopes for approximately a decade at the University of Minnesota School of Medicine at Duluth. This achievement was made possible by the establishment of a multimedia-based laboratory that included archiving of static labeled histological images. The images were placed originally on videodiscs and subsequently on CD-ROMs (Downing, 1995, and personal communication). Virtual slides and virtual microscopes have been integrated successfully into the laboratories in histology along with microscope glass slides and light microscopes at the University of Iowa (Harris et al., 2001; Heidger et al., 2002). To our knowledge, ours is the first report of a successful complete transition from the use of microscope glass slides and microscopes to the combined use of static labeled images and virtual slides and virtual microscopes for the teaching of laboratory sessions in medical histology courses in North America.

Acknowledgements

We thank the following faculty members who were involved with the changing of the teaching of Medical Histology at the University of South Carolina after 1988: James R. Augustine, Fredric R. Boockfor, Gary T. Campbell, James A. Hightower, and D. Louise Odor. We also thank Christopher M. Prince of Petro Image, LLC, for his expertise in the production of the virtual slides.

Ancillary