Virtual Microscopes in Podiatric Medical Education

Abstract

In many medical schools, microscopes are being replaced as teaching tools by computers with software that emulates the use of a light microscope. This article chronicles the adoption of “virtual microscopes” by a podiatric medical school and presents the results of educational research on the effectiveness of this adoption in a histology course. If the trend toward virtual microscopy in education continues, many 21st-century physicians will not be trained to operate a light microscope. The replacement of old technologies by new is discussed. The fundamental question is whether all podiatric physicians should be trained in the use of a particular tool or only those who are likely to use it in their own practice.

The advent of powerful image scanners has allowed creation of the virtual microscope, which emulates a light microscope on a computer [1]. Virtual microscopes have recently been used as teaching tools in subjects that have traditionally used light microscopes, such as geology and histology. Several medical universities, such as the University of Iowa and the University of California at Los Angeles, have reported great success in teaching histology or pathology laboratory courses using virtual microscopes and have either partially or totally eliminated the use of light microscopes and glass slides [2–5]. Others have used the virtual microscope for teaching histopathology skills to physicians in continuing medical education sessions [6^,7].

Traditional glass slides of sections of histologic or histopathologic specimens can be scanned through a microscope as digital images and stored electronically. Because the sections are scanned at higher microscope powers, the field becomes narrower, but imaging software stitches together the small fields into nearly seamless larger images. The digital images, termed virtual slides, are stored on large servers. Appropriate hardware and processing equipment allow more than 100 microcomputers, linked in a network, to simultaneously access virtual slides from the virtual slide box on the server(s). Many universities have put scanned histologic specimens onto the Web or onto compact discs as static images, but this format does not emulate a microscope. Commercial firms and universities have developed software to scan and zoom the fields, thus emulating moving the microscope stage and changing the objective lenses, respectively. The term virtual microscope is used here to describe computers that have this capability. Some software packages also have ancillary tools, such as fine focusing, pointers, labelers, and measuring tools and the ability to simultaneously display low- and high-power images of the same specimen.

Rosalind Franklin University of Medicine and Science had taught medical histology with light microscopes and glass slides for many years, and the Pathology Department of Chicago Medical School was one of the pioneers of using static computer images in the laboratory. In 2003, Rosalind Franklin University decided to use the histology laboratory space for other purposes, and in the process of searching for new space, the histology teaching faculty proposed in early 2004 the use of the student computer room as a virtual laboratory. Chicago Medical School and Dr. William M. Scholl College of Podiatric Medicine contracted with Bacus Laboratories Inc (Lombard, Illinois) to scan the best glass slides in the collections of both schools. In the summer of 2004, 140 new computers were installed in the student computer laboratory, and the Bacus software was installed in each machine. All of the computers were networked to a server that contained the scanned images. Laboratory manuals were rewritten to accommodate the virtual microscopes. Because of the haste to complete construction of the laboratory space, the medical and graduate students began their histology course in August 2004 with traditional microscopes and glass slides and then converted to the virtual microscopes after 2 weeks. The podiatric medical students used the virtual microscopes throughout their structure and function (histology and physiology) courses, which also included pathology assistant students (one-fifth of the class). When the students in each course finished their final practical examination, they were asked to complete a survey specifically regarding the laboratory portion of the course and emphasizing the change to virtual microscopes and slides. All of the laboratory teaching faculty members were interviewed soon after the end of both courses.

The purpose of this article is to report on the results of the surveys, to discuss the advantages and disadvantages of using virtual microscopy, to discuss the relative merits of teaching process versus concepts, and to estimate the need for light microscopes in the medical professions. To our knowledge, this is the first study of the use of virtual microscopy in podiatric medical schools.

Methods

Rosalind Franklin University owned more than 100 light microscopes and boxes of glass slides. Each school chose the best slides in their collections and had them scanned by Bacus and installed on the server. Students accessed the program in the student computer laboratory by entering the password-protected Rosalind Franklin University Web site, then opening the software developed by Bacus, the WebSlide Server. They then had a choice of several virtual slide boxes: the collections from Chicago Medical School and Scholl College of Podiatric Medicine and other collections that Bacus makes available to institutions that contract with them. Once a slide has been selected, the student has the choice of eight virtual objective lenses, with a magnification range of ×1.6 to ×400. Students emulate moving a glass slide across a microscope stage by using the computer mouse and a “hand” screen tool. Other tools were a linear tape measure (helpful for comparing the radii of blood cells in a smear), a tool that allows the student to draw a perimeter around an object and then calculates the area, and a pointer. Innovative students also learned how to export the images to graphics or presentation programs and to perform additional tasks, such as adding labels. Although most of the glass slides chosen for scanning were the best in the collections, occasionally some specimens were sectioned thickly. The Bacus software allows emulation of the fine focus on a light microscope so that students can use keyboard commands to focus up and down, within a limited range, through a specimen. Because this feature uses more memory, it was used on only some of the slides. Students had three options for browsers for viewing the images: ActiveX, Java, and three-panel. The latter simultaneously shows high-, medium-, and low-power images of the same slide on three panels of the computer screen. Although each student had his or her own computer, many students congregated around one computer in groups of two to five and worked cooperatively. The teaching laboratories were staffed with the same student-faculty ratio as in previous years. In the past, instructors had used multiheaded microscopes or microscopes with charge-coupled device cameras connected to monitors to demonstrate special features to groups. In the virtual laboratory, instructors simply chose one student’s computer and gathered a group around it.

Practical laboratory examinations were held in the computer room using a set of images from the virtual slide box. The images were imported into PowerPoint (Microsoft Corp, Redmond, Washington), labeled, captioned, and then converted into an Adobe PDF file that the students could not alter. The file was stored on the server under a special password. Traditional practical examinations allowed students a fixed amount of time at each microscope station, and return to previous stations was prohibited. In the virtual laboratory examination, the overall time was limited to 1 hour, but students could spend as much time as needed on each slide, and return to previous slides was allowed. Students wrote their answers on papers, which were graded by hand; in the future we will attempt to use online answer sheets and a secure browser [8].

The survey was written with Likert-type choices, was submitted to the Rosalind Franklin University institutional review board for approval, and was administered just after the students finished their final laboratory practical examination. Return rates were greater than 90% for both classes, and the results for each question were counted and are expressed as percentages. The written comments were read for recurrent themes, and tallies were made for each theme. Nine faculty laboratory instructors and one graduate student teaching assistant were interviewed by the principal author using a structured format [9]. Interview questions were clustered around four issues: convenience, teaching effectiveness, examinations, and professional preparation.

Results

The surveys included five information-gathering questions with yes-or-no answers, 15 attitudinal questions with four Likert-type preferences (strongly agree, agree, disagree, and strongly disagree), one question estimating their grade in laboratory, and two questions for written comments that asked what the best feature of the virtual microscopy laboratory was and what needed to be improved. Table 1 displays the results for selected questions from the survey of podiatric medical students plus pathology assistant students, and Table 2 displays the results from the nearly identical survey of medical students. Approximately 85% of the medical students and 97% of the podiatric medical students agreed or strongly agreed that computer-based virtual microscopy enhanced their understanding of histology. In general, the responses were similar in the two groups, and the differences likely reflect the different format of the two courses (eg, the medical course has more variation in the slide choices).

The comments written in response to the question about the best feature of the virtual microscopy laboratory were very favorable, with most of the students commenting on the ease of use and the convenience of access whenever the computer center was open. The written comments about necessary changes in the future focused mainly on the need for more memory in the computer server when many students try to simultaneously access the same slide.

The responses from the interviews with the instructors were read for common themes. Although many of the instructors’ comments were similar, each person had a salient observation. In addition to the advantages of the virtual microscope for students, there are also benefits and conveniences for teachers. For example, laboratory instructors found that group teaching was much easier with virtual microscopes. On the other hand, some instructors felt that the virtual slide box encouraged more passive learning than when students had to actively find a particular image feature on glass slides. The advantages and disadvantages of virtual microscopy for teaching histology laboratory courses are summarized in Table 3.

Discussion

Virtual microscopy was preferred to traditional microscopy by students and teaching faculty members, a finding that is common among universities that have evaluated the change [2–4,8]. Most students were swayed by convenience, although convenience and learning issues are often intertwined. For example, it is more convenient for most people to view images on a computer screen than through the eyepiece of a light microscope. However, it is also an advantage for learning because laboratory instructors or peer students can simultaneously see the same image, which facilitates group learning. Group teaching can be performed with light microscopes with multiple heads or through a microscope camera connected to a monitor, but image quality suffers. Another example is the convenience of the virtual slide box versus the many boxes of glass slides. The glass slides varied because of staining and sectioning differences. In the older method, students occasionally had to look at the slides of their peers to ensure that they had seen all of the variations of any particular slide because any version might be chosen for a practical examination. The homogeneous virtual slide box also facilitates learning because less variety need be committed to memory, allowing students to concentrate on the distinguishing features of an image. The counterargument is that only perfectly stained and sectioned images are chosen for virtual slide boxes, print atlases, course notes, static electronic images on compact discs or the Web, and, finally, examinations. Thus students merely memorize archetypes of histology or pathology and do not learn all of the information inherent in variation.

The concern about rote memorization was heightened when students were examined in the laboratory. Students were tested by means of the practical method at midterm and at the end of the term. Changes from the past were that the images were selected from the virtual slide box, the students sat at a computer instead of moving from microscope station to station, time at each image was not limited, and students could return to an image for a second look. For both examinations and in both courses, the mean laboratory grades were significantly higher than in previous years, although little had changed in the lecture portion of either course. The improved performance was discussed in interviews with the instructors. Most instructors felt that part of the improvement was due to the reduction in stress from not having to move from one station to another, refocus the microscope for their eyes, and identify the specimen in a fixed amount of time. The students were familiar with the station-to-station practical examination from the gross anatomy course, and many agreed that the new method used in the histology course was preferable. The instructors felt that the second reason for higher laboratory grades was the homogeneity of the virtual slide box; students could more readily memorize a smaller set of images for the recognition questions. This was less true for the medical school course, which uses more functional than recognition questions. Some instructors felt that students’ awareness of this experiment (a Hawthorne effect) accounted for a very small part of the rise in grades. Incidentally, cheating is less likely than when students are constantly moving around in the traditional practical examination. Attempts to retrieve other information on the computer can be blocked with a secure browser [8]. Nonetheless, modern students are very skilled with computers, so medical schools should use precautions as necessary. Finally, professors of histology and pathology should ask whether the practical examination has any role in the future. If students can take examinations in secure computer laboratories, then faculty who prepare examinations need not distinguish lecture content from laboratory content.

Administrators were concerned about the costs of the conversion, and there were significant initial costs. There will be some continuing costs for scanning a few additional slides, for the annual software-rental agreement, and for upgrade of the servers. There are also indirect annual costs associated with using Rosalind Franklin University’s information-technology personnel. Conversely, there will be savings from not having to repair and clean the microscopes annually, not refiling glass slides and replacing broken slides, and not maintaining a large laboratory space that was seldom used at other times of the year. The computer laboratory, on the other hand, is open and used year-round by students for multiple purposes. Rosalind Franklin University also gained space from not having to store the microscopes and some revenue from the sale or donation of most of the microscopes.

It was originally thought that conversion to virtual microscopy was a singular event, but the conversion actually brings two changes to most teaching laboratories: the virtual microscope itself and the virtual slide box, which contains identical images for all of the students and instructors. There are other methods for homogenizing a slide box without using a virtual microscope, but none are very practical for a large teaching laboratory, and more variety could be added to a virtual slide box by scanning more sections of each specimen or more specimens. Nonetheless, most of the medical schools that convert to the virtual microscope also convert to the homogeneous virtual slide box because of cost and convenience. The fundamental learning issue, then, is whether doctors-in-training in the 21st century need to examine the histology of tissue specimens as scrupulously as in the past. For medical students, knowledge of normal histology is most important for improving knowledge of organ function. For medical practitioners, knowledge of normal histology is useful for comparison with histopathology. If normal morphology can be learned as an archetype, and pathology as deviations from it, then perhaps learning normalcy from a homogeneous slide box will suffice. If this postulate is accepted, it then begs the question of whether pathology can likewise be learned from archetypes, with a few variations.

An argument can be made that since enactment of the Clinical Laboratory Improvement Amendments of 1988, few physicians or podiatric physicians have used a microscope, and fewer still own one. Most doctors send specimens to commercial histology laboratories, which do all of the processing, microscopy, and interpretation, and then the doctor is sent a written report. The commercial histology laboratories can send the doctor an electronic image from a scanned slide if requested, but often the written report suffices. In case the written report is equivocal, the doctor always has the option to send the specimen to another laboratory or to the Armed Forces pathology laboratory. In one sense, doctors are using virtual slides from a specimen that has been homogenized through the expert processing of the commercial laboratories, and filtered through the expert opinion of the report writers. One argument is that such laboratories are susceptible to mistakes and that the doctor should be a manager who should know how to examine the slides with the same skill as the laboratory and thus has the responsibility for the laboratory’s conclusions. The counterargument is that modern doctors cannot be experts in all of the techniques in medicine and that they must delegate some responsibility to others on the health-care team. Physicians operating in this mode may learn some procedures or skills from archetypal examples and leave the details to others, although the doctors assume ultimate responsibility for the patient’s care. Medical doctors still receive a general education in medicine during their usual 4 years of school. Thus, for example, future dermatologists and psychiatrists receive didactic and clinical training in obstetrics. The analogous question is whether all physicians should be trained in histology to the same degree as pathologists, oncologists, and others. Should those who opt for these specialties acquire more refined microscopy skills in later training? Should physicians who intend to operate a histopathology laboratory in accordance with the Clinical Laboratory Improvement Amendments of 1988 also learn more histology later? Another argument is that physicians who will practice in rural or Third World settings must learn to perform procedures that are not locally available. Should educators train all physicians to a level that encompasses these exceptional cases, or should the exceptions be encouraged to train themselves to have greater skill with a light microscope?

The issue of convenience versus learning also arises for the virtual microscope itself. The software that allows emulation of a light microscope (scanning the field, changing magnification, and focusing) is an improvement over static electronic images, but it is more expensive. Furthermore, the virtual microscopes of today cannot emulate all of the nuances of optical microscopy, such as using the iris diaphragm for greater contrast (although many of those nuanced techniques were necessary because of poor specimen preparation). The survey described here found that more than 90% of the students in 2005 had used a microscope earlier in their education. If high schools and colleges eschew microscopes for electronic learning modes, then the next generation of students will arrive at medical school completely naive about operating microscopes. For that generation, will emulation be necessary at all? The fundamental question is whether future doctors need to be experts in the operation of a tool or simply be able to interpret its results.

These questions remain unanswered here because the debate is ongoing, and medical schools have been switching to total virtual microscopy for teaching histology and pathology laboratory courses. In fact, a session on “Virtual Microscopy in Teaching Histology” was included in the program of a national meeting of the Federation of Societies of Experimental Biology in San Diego in April 2005. In June 2005, a book [10] was published on the topic of virtual microscopy. On the other hand, podiatric medical students have already chosen their specialty on entry into 4 years of medical school, and their training as generalists is limited. Very few podiatric physicians process their own tissue specimens anymore (Nancy Parsley, DPM, American Podiatric Medical Association, e-mail communication, 2005). Learning histology and pathology from archetypes will serve future podiatric physicians well, and those who need more will have opportunities for advanced microscopy training in postgraduate programs, in certification programs, or by informal training and self-initiated learning.

The surveys were written more to elicit students’ personal impressions than to attempt an assessment of teaching effectiveness. Also, because the students were aware that they were participating in a new program, there is a Hawthorne effect in the students’ self-assessment of how much more they learned using virtual microscopy. Other assessment measures will be followed later in the careers of this student cohort, such as results of the histology portions of board examinations and performance in second-year pathology courses.

Conclusion

Students and instructors much preferred the virtual microscope and slide box to the traditional methods of learning laboratory histology. It is hoped that the energy gained from the ease and convenience of virtual microscopy was invested in other learning endeavors in medical school. However, there do seem to be real learning advantages for general and podiatric physicians. If traditional microscopes are removed from undergraduate medical training, then graduate medical educators must ensure that those specialists who need training in light microscopy receive it.

Table 1. Selected Survey Results for 102 Podiatric Medical and Pathology Assistant Students.

Table 1. Selected Survey Results for 102 Podiatric Medical and Pathology Assistant Students.

Table 2. Selected Survey Results for 154 Medical Students.

Table 2. Selected Survey Results for 154 Medical Students.

Table 3. Advantages and Disadvantages of Virtual Microscopy.

Table 3. Advantages and Disadvantages of Virtual Microscopy.