As imaging has been adopted in the modern medical education, it has benefited from the concurrent development of technologies that have allowed the material to be presented electronically. One of the technologies with the greatest impact has been the Internet [9]. The Internet has increasingly been utilized as an educational tool due to its ability to provide a large volume of educational material in a single, readily-accessible location as well as permitting flexibility in the material format. Images, text, interactive quizzes, and videos can be integrated seamlessly into a comprehensive educational resource. While these capabilities are largely beneficial, they can have a negative effect on the efficacy of the educational resource if the material is poorly organized. For example, if students are confronted with a vast amount of information on the Internet that is presented in a non-user-friendly format, they are likely to either not utilize the resource or not benefit from its use.

One proposed solution for managing Internet material in an organized, user-friendly format is the implementation of a web portal. A web portal is a site that serves as a single point of access to information collected from different sources and presented in multiple formats. Some common features of portals include personalized navigation, for example, “quick links” to frequently accessed information pages, directory-based information structure, community-building tools such as chatrooms, bulletin boards, and emailing lists, user authentication (log in and password), and subject-specific search functionality. Web portals have recently gained in popularity as they have been successfully implemented in a number of clinical settings, such as nursing [10], mental health [11], government disaster preparation [12], patient diabetes information [13], and the MedEdPORTAL implemented by The Association of American Medical Colleges [14]. The general consensus from these studies has been that web portals require initial planning and subsequent maintenance in order to remain relevant and effective but when designed appropriately, this online educational format is robust, easily accessible, and associated with excellent outcomes. While there is limited information regarding the concept of utilizing a portal for medical education purposes, it is reasonable to consider that these benefits recognized in clinically focused web portals would translate effectively for a medical-education-related web portal.

A survey from the previous year with student feedback from initial efforts in developing Radiology teaching files in a web-based format for Anatomy was reviewed. The team of attending, residents, and medical students working on the second iteration of development used this feedback to identify several possible areas for improvement with the goal for improved user friendliness and increased utilization of the material. Some of the key changes included integration of the material into a web portal, development of Lab and Study Companion components, and a set of organized links to access the material rather than requiring the user to type in search criteria. The list of features that were created and the rational for their implementation are provided in Table 1.

Joomla 1.5 (Open Source Matters, Inc, New York, NY) was utilized as a content management system for creating the web content. Joomla is an open-source application that is freely available on the Internet. It provided the ability to organize and keep track of all content as well as constantly update cases and documents without republishing the web page.

In order to expand the content that would be offered via the web portal, a search for relevant images within the past year was conducted using the Johns Hopkins Hospital Emageon Ultravisual (AMICAS, Inc, Boston, MA) picture archiving and communication system (PACS). There was no copyright protection for these images developed in this iteration of the site development. Although the site was password protected, there was no functionality limiting copying of images from the site by students. This issue will be further addressed in future iterations of the web site development. The effort to obtain several relevant images for each anatomic site was largely in response to student feedback from the prior year that suggested additional imaging teaching files would be beneficial. In order to expand the MIRC content already available, the goal of case acquisition this year was directed towards finding CT and MRI cases where multiple slices could be viewed for each case. Similar to the existing teaching files, the new cases included normal anatomy as well as cases with simple pathology. The cases were initially identified by two of the authors and subsequently reviewed by the senior authors before incorporation into teaching cases. Once selected, the image sets were saved in standard tagged image file format and animated with Macromedia Flash using an open-source student PACS module created by students and residents at the University of Medicine and Dentistry of New Jersey. The new modules allowed users to scroll through images, zoom, and pan while also interacting with labeled structures within the image sets. The decision to utilize Flash was impart based on familiarity of the product by the student developers. In addition, while evaluation of MIRC demonstrated excellent static image teaching file development, it had less robust capabilities compared to the Joomla/Flash combination option for developing more dynamic teaching files with scrolling and highlighting.

The new student PACS modules were combined with last year's MIRC cases to create teaching files that could be accessed during lecture by the instructors and by the anatomy students for review during and following anatomy lectures. MIRC files which were created last year were incorporated into the web site as thumbnails with hyperlinks to the original content, while new thumbnails were created to hyperlink to the students PACS cases. The web portal was designed such that all teaching files were organized by anatomical region. These anatomical regions were defined based on the existing syllabus for the anatomy course. The files were further separated into tutorial files and quiz files. In total, over 100 teaching files were provided (Tables ​(Tables22 and ​and33).

In order to create interactive teaching files, relevant anatomy was identified and highlighted on each slice of an image set. A set of questions or relevant teaching points were then linked to each structure so that if the structure was selected on any slice where it could be seen, the questions or teaching points would then appear on the right side of the screen. Figure 1 displays a sample tutorial case where the arch of the aorta was highlighted and relevant information was displayed for that structure. In Figure 2, a quiz case is displayed where the ascending aorta has been highlighted prompting a multiple choice question related to this structure. For each of the tutorials, there was a page of instructional text, with one associated image set as demonstrated in Figures ​Figures11 and ​and22 and a corresponding set of relevant multiple choice questions.

The web portal was introduced to the medical students on the first day of their anatomy class. The rational for presenting the portal on the first day was to provide students a roadmap to optimize use of Radiology to learn Anatomy and to begin to get them comfortable with the basic layout of the web site. The web address was provided and basic navigation through the site was reviewed. The students were informed that the homepage would be updated a few days prior to each radiology lecture to include pertinent cases (Figure 3). They were required to look over those selected cases prior to lecture and advised to look over the additional cases found in the corresponding anatomical link for further review.

Each radiology lecture was divided into two one-hour segments. During the first hour, the lecturer presented a PowerPoint (Microsoft Corporation, Redmond, WA) lecture on material corresponding to the gross dissection scheduled for that day. The students were then given a five-minute break and asked to close all laptops before the second session began. Next, the lecturer broke the classroom up into small groups based on where the students were sitting and quizzed them on each of the cases found on the homepage of the web portal for that day. This process was repeated for each of the radiology-based lectures so that the students were aware of the expectation that they should be familiar with the corresponding cases found on the web portal homepage.

Following the completion of the Gross Anatomy course (Fall 2010), a fourteen question, web-based survey was distributed to the Anatomy students. The survey was similar to the one distributed to the first-year medical students the previous year (Fall 2009) [15]. Figure 6 provides a complete list of the questions on the survey. In order to evaluate the impact of the web portal compared to the previously available teaching files, the results of the survey from the current year were compared to the results from the previous year.

Compared to the previous electronic teaching files, the new content received better evaluations for user friendliness and usefulness (Table 4). The mean score for usefulness of the web portal content was 3.3 points (helpful to very helpful) compared to 1.5 points (not helpful to somewhat helpful) for the previous material (P < 0.001). When excluding the 0 point scores of the students who did not use the online teaching files last year, the mean score improved from 1.5 to 2.6 points (somewhat helpful to helpful). However, the 2.6 points remained statistically less than the 3.3 point rating for the web portal (P < 0.001). Similarly, the new web material was rated as being more user friendly 3.1 points (good to excellent) versus 2.3 points (satisfactory to good) even after correcting again for the students who did not use the online content (P < 0.001).

The primary reason students utilized the resources was to review material that was going to be tested (Figure 4). There were 52 (60%) students who ranked this as the number one reason for accessing the teaching files, and 67 (77%) students ranked it as either the no. 1 or no. 2 reason they used the web material. The number of students who indicated that the teaching files were used because they helped to better understand anatomical relationships was only slightly less with 66 (76%) students ranking this option as either their no.1 or no. 2 choice. There were few students who indicated that their primary reason for using the teaching files was because they were free. However, many students ranked this as their second or third reason for utilizing the site, and this was the third highest ranked choice overall.

The numbers of students in each of the test score categories were 2 students with average scores 61 to 70, 11 students with scores 71 to 80, 37 students with scores 81 to 90, and 37 students with scores 91 to 100. There appeared to be a trend for the highest scoring students (91 to 100 averages) to utilize the web portal more frequently than the 71-to-80 and 81-to-90 groups (Figure 5). However, there were insufficient numbers of students in the various test score groups to perform a chi-square analysis with any sufficient degree of certainty.

There were no students who completed the survey that did not use the radiology-based teaching materials available on the web site. This finding was consistent with required prelecture study files and subsequent review files that were provided as study aids for similar material that would be tested. The web server traffic pattern demonstrated a baseline of 30 to 40 hits per day. The number of hits increased to over 100 during the two to three days prior to an exam.

One of the factors that may have facilitated the development of a useful tool for Gross Anatomy students is that some of the authors are medical students or are only a few years removed from taking the course themselves and can relate to what material may be useful for Gross Anatomy study. This hypothesis is supported by a recent study by Rosenbaum et al. [16]. They evaluated medical student involvement in the development of a website to act as each individual medical student's homepage. They created a web portal that provided students with access to course material, evaluations, academic information, and community assets. Based on greater than 80% positive feedback regarding the web portal, they recommended that other medical schools that are creating and expanding digital resources should solicit the valuable input and perspective of medical students.

It was and will be difficult to analyze the effect of the web portal on student performance since it was felt to be inappropriate to withhold access from half the class for a controlled comparison. Our students are highly motivated and sophisticated learners, as well as “digital natives” accustomed to using web resources; it was felt their comments and satisfaction scores were legitimate metrics of utility or the resource. There also continue to be changes in the overall Gross Anatomy course structure and student testing reflecting as our medical school's new curriculum moves through its second year, further complicating comparison of test scores or grades from the initial year of web site usage to the second year with the improved/modified web material.

Although the primary objective for the present study was to create user-friendly teaching files for anatomy students, the ultimate goal of the Radiology research team working on this project was to create the foundation for a medical student and resident Radiology education web portal. The concept for the web portal was derived from previous studies at our institution that have highlighted the benefit of longitudinal directed teaching [7, 17]. For example, Feigin et al. reported that recall and retention toward the end of medical school was facilitated by providing preclinical instruction followed by review of the material later in school [7]. They reported that senior students benefited from previously received preclinical radiology training as they were able to improve their score on an anatomy quiz from an average of 4.42 (standard deviation 1.34) to a score of 8.65 (standard deviation 1.24) only three weeks later. The goal for the web portal in the present study is to provide a similar benefit but in an electronic format. The web portal design updates that have now been included in an effort to meet this goal are components for (1) imaging modules to be integrated into the basic clerkships (PRECEDE Modules for third year medical students), (2) lectures from The Johns Hopkins Basic Radiology Elective (for third- and fourth- year medical students), and (3) Johns Hopkins Radiology Resident Joint Procedure notes (for residents). For those students interested in pursuing a career in radiology or just looking for residency match advice, a residency application guide (APPS of STEEL) created by the senior author was also included. Finally, links to other recommended anatomy and radiology resources were incorporated into the site to allow students and residents to further their interest in the field.

While all of this functionality of the portal was customized for a single institution, the authors are assessing possible collaborations with other institutions. There is a potential to provide a more universal product. This product could potentially be provided to international colleagues who may not have the funds to develop a similar resource. To provide this outreach, it is conceivable that such a product could linked to be included under the umbrella of a larger entity with greater outreach such as the MedEdPORTAL.

The students indicated that the number one reason for accessing the files was to prepare for the material possibly being on the test. Although this may have been the primary incentive, it is important to note that regardless of the reason the students accessed the material initially, there were a large number of students who subsequently found the material to be user friendly and helpful in learning to identify anatomic structures. Furthermore, it is reasonable to expect students who are required to utilized material that will be tested to be more hypercritical of the material. As such, the improvements in reported user-friendliness of the electronic material are arguably more substantial. Of interest, students constantly came forward or emailed us during the semester to make constructive suggestions or to inquire about joining the project, also implying engagement. These findings suggest that Educators should recognize that by informing students that material will be on their tests, a majority if not all of the student body will review the resource. The goal of the Educator, however, should not only be to have students utilize the resource but to also provide material that is user friendly and provides optimal educational value. If Educators are considering a digital format for distributing material, a structured web site can be a tool to provide easy access, in a user friendly format.

Similar to other questionnaire-based studies, one of the limitations of the present study is the number of students who do not participate. As previously noted, the response rate of 73% for the present study was similar to the rate for the survey from the previous year. This response rate is approximately one standard deviation above the normal response rates reported in the literature [18]. While it is uncertain how the remaining 27% of students would have answered the survey questions, we believe that based on the reported demographics and test scores that the study population was a fair representation of the class on a whole. Another limitation is that the survey was not formally validated prior to online implementation. However, students did not report problems when completing the survey, and standard survey functionality provided by the vendor was utilized to ensure students completed all required questions. One other limitation was the accuracy of the self-reported test scores. However, the range of scores reported appeared to reflect the class actual performance and with no identifiers for the responses the students had no incentive to not report their true performance.