Medical Image Resource Center–making electronic teaching files from PACS

A picture archive and communications system (PACS) is a rich source of images and data suitable for creating electronic teaching files (ETF). However, the potential for PACS to support nonclinical applications has not been fully realized: at present there is no mechanism for PACS to identify and store teaching files; neither is there a standardized method for sharing such teaching images. The Medical Image Resource Center (MIRC) is a new central image repository that defines standards for data exchange among different centers. We developed an ETF server that retrieves digital imaging and communication in medicine (DICOM) images from PACS, and enables users to create teaching files that conform to the new MIRC schema. We test-populated our ETF server with illustrative images from the clinical case load of the National Neuroscience Institute, Singapore. Together, PACS and MIRC have the potential to benefit radiology teaching and research.     

A Novel Application of the MIRC Repository in Medical Education

Medical students on the radiology elective in our institution create electronic presentations to present to each other as part of the requirements for the rotation. Access was given to previous students presentations via the web-based system, Medical Imaging Resource Center (MIRC) project, created and supported by the Radiological Society of North America (RSNA). RadPix Power 2 MIRC (Weadock Software, LLC, Ann Arbor, MI) software converted the Microsoft PowerPoint (Redmond, WA) presentations to a MIRC-compatible format. The textual information on each slide is searchable across the entire MIRC database. Future students will be able to benefit from the work of their predecessors.     

Establishing a Web-Based DICOM Teaching File Authoring Tool Using Open-Source Public Software

Online teaching files are an important source of educational and referential materials in the radiology community. The commonly used Digital Imaging and Communications in Medicine (DICOM) file format of the radiology community is not natively supported by common Web browsers. The ability of the Web server to convert and parse DICOM is important when the DICOM-converting tools are not available. In this paper, we describe our approach to develop a Web-based teaching file authoring tool. Our server is built using Apache Web server running on FreeBSD operating system. The dynamic page content is produced by Hypertext Preprocessor (PHP). Digital Imaging and Communications in Medicine images are converted by ImageMagick into Joint Photographic Experts Group (JPEG) format. Digital Imaging and Communications in Medicine attributes are parsed by dicom3tools and stored in PostgreSQL database. Using free software available from the Internet, we build a Web service that allows radiologists to create their own online teaching file cases with a common Web browser.     

Implementing a MIRC Query Interface for a Database Driven Teaching File

This paper describes the authors experience with integrating an existing database-driven teaching file with the RSNA (Radiological Society of North America) Medical Imaging Resource Center (MIRC). MIRC is the product of an RSNA-sponsored initiative to enable medical institutions to share their electronic medical content (images, text, and multimedia) by creating a distributed repository accessible from the Internet. An existing database-driven teaching file, developed by the authors and used extensively by the University of California San Francisco (UCSF) Department of Radiology since 1998, was retrofitted to include an interface for handling broadcast queries initiated by a MIRC query service. These queries take place through the exchange of XML documents via HTTP. After all the storage services have responded, the results are collated by the query service and presented to the user. The teaching file and MIRC interface were developed using the 4th Dimension Relational Database Management System (RDBMS). The integration process primarily involved mapping the MIRCdocument schema to the teaching files schema, translating the actual MIRC query into the internal query language of the database and extending the access control mechanisms of the teaching file to allow public access. A working implementation of the interface required only 3 days of development time, with refinements taking place over several months. Interface development was greatly aided by MIRCs use of well-established Internet standards. This project has demonstrated the feasibility of implementing a MIRC interface on an existing teaching file server.     

Electronic teaching files and continuing professional development in radiology

Education in diagnostic radiology employs medical images extensively, and case-based teaching files of actual patients are useful to illustrate pertinent teaching points. In the era of digital radiology, there is great potential to use the ready source of patient material from Picture Archive and Communication Systems (PACS) and initiatives such as Medical Imaging Resource Center (MIRC) on the World Wide Web for teaching and for Continuing Professional Development (CPD). As mandatory CPD becomes the reality in medical practice, computerised solutions that support the creation of electronic teaching files in the midst of busy clinical workflow would be very valuable. This paper will explore the features of image-based CPD, the various ways in which medical images can be used for self directed learning and the challenges that face the radiology profession.     

Distributed radiology clerkship for the core clinical year of medical school.

OBJECTIVE: The central role that diagnostic radiology has in the modern practice of medicine has not always been reflected in radiology's place in the curriculum. We developed a new radiology clerkship for undergraduate medical students during their core clinical year that was supported by Web technology. The assumptions underlying the design of the clerkship were that radiology is best learned from radiologists and that students are most receptive to learning radiology when it is related to concurrent patient care experiences. DESCRIPTION: Beginning in May 2000, a required radiology clerkship experience was incorporated into the core clinical year at Wake Forest University School of Medicine. The core clinical year was organized into three 16-week blocks of clerkships. Two or four independent half-day radiology tutorial sessions were included with each clerkship block, and attended by all students in the block (approximately 35 students), regardless of their specific clerkship assignments. There were ten different radiology tutorials, each given three times during the year as students rotated through the clerkship blocks. Thus, each student attended a radiology tutorial session every four to eight weeks during the year. The topics covered during the tutorials were correlated with the content of the clerkship blocks and included adult and pediatric chest radiology, adult and pediatric abdominal radiology, body CT, neuroradiology, obstetric ultrasound, gynecologic ultrasound, osteoporosis, adult and pediatric fractures, mammography, and cervical spine trauma. The tutorials included pre- and post-test, lectures, case presentations, and sometimes tours of the radiology department. The educational emphasis was on pragmatic case-based learning exercises, development of verbal and visual vocabulary, and learning when and where to seek more information. To provide continuity and organization, Web-based curriculum materials were designed and implemented as a component of the clerkship. The home page of the Web site provided the schedule, faculty names, attendance and grading policies, course overview, and links to individual tutorials. The pages for individual tutorials included educational objectives, glossary of radiology terminology relevant to the subject, lecture slides and handouts, and teaching cases. All students had laptop computers and access to the academic network, but did not use them during the actual tutorial sessions. DISCUSSION: Implementation of the radiology clerkship required extensive negotiation with directors of other clerkships so that students could be released from their other responsibilities in order to attend the radiology tutorials. The radiology clerkship format has proven to be complex in its administration, with faculty and students on different schedules commuting to the radiology lecture hall from various locations. Extensive use of e-mail and communication via the Web site have been instrumental in reminding faculty and students of upcoming sessions. Preliminary evaluations have indicated that students liked the radiology sessions and learned a great deal, but disliked the scheduling and the lack of continuity. An evaluation of the curriculum and its components is ongoing.     

Expression of cyclooxygenase-2, c-kit, progesterone and estrogen receptors in uterine smooth muscle tumors: differential diagnosis

We examined the expression pattern of cyclooxygenase-2 (COX-2) and c-kit in uterine smooth muscle neoplasms and tried to determine the role of these markers in differential diagnosis. Archival tissue from 64 patients with uterine smooth muscle neoplasms (20 leiomyomas (LMs), 22 atypical leiomyomas (ALMs), and 22 leiomyosarcomas (LMSs)) was immunostained with antibodies against estrogen (ER) and progesterone receptors (PR), COX-2 and c-kit. 7 of 20 LM cases and 5 of 22 ALM cases were immunopositive for COX-2, whereas none of the LMS cases stained immunopositive (p< or =0.05). 4 of 20 LM cases and 5 of 22 ALM cases were immunopositive for c-kit, whereas 15 of 22 LMS cases showed c-kit immunopositivity (p< or =0.05). In conclusion, very few LMs and ALMs show COX-2 immunopositivity. LMSs usually do not express COX-2. COX-2 expression in smooth muscle tumors is not a prominent feature. Therefore, COX-2 inhibitors may not be useful in LMS therapy. C-kit was significantly expressed in uterine LMSs.     

Use of X-chromosome inactivation pattern to determine the clonal origins of uterine leiomyoma and leiomyosarcoma

Uterine leiomyomas (LMs) and leiomyosarcomas (LMSs), both of smooth muscle origin, sometimes coexist in the same uterus. Little genetic evidence exists concerning the developmental relationship between LM and LMS. Using the X-chromosome inactivation pattern of the human androgen receptor gene, we examined the clonality of LM and LMS. Of the 24 patients with LM, 21 had multiple neoplasms; all were clonal and individual LMs derived from separate clones. Of the 20 patients with LMS, 6 exhibited multiple tumors in the uterus, and 4 of these individuals also harbored coexisting uterine LMs. We found all LMSs to be clonal. Separate tumors showed identical pattern of X inactivation in 4 patients, and in 2 other individuals, multiple LMSs developed from independent clones. Among the 4 patients with LMS and coexisting LM, 3 showed the same pattern of X inactivation in LMS and the adjacent LM. In 2 of the 3 patients, the tumor also exhibited a morphological transition between benign cells in LM and malignant cells in LMS, supporting the possibility of transformation from LM to LMS. One patient displayed different clones in LMS and the coexisting LM, indicating their independent origins. We conclude that (i) both LM and LMS are clonal; (ii) different nodules in multiple LM are of independent origins; (iii) multiple lesions of LMS may be either monoclonal or multiclonal; (iv) most LMSs are solitary lesions and are most likely de novo, but an individual LM may undergo "malignant transformation" to a LMS; and (v) some LMSs and coexisting LMs are of independent origins.     

Incorporation of a Formalized Emergency Radiology Curriculum to Facilitate Population of a MIRC-based Digital Teaching File

Teaching files are integral to radiological training. Digital Imaging and Communication in Medicine compatible digital radiological data and technological advances have made digital teaching files a desirable way to preserve and share representative and/or unusual cases for training purposes. The Medical Imaging Resource Community (MIRC) system developed by the Radiological Society of North America (RSNA) is a robust multi-platform digital teaching file implementation that is freely available. An emergency radiology training curriculum developed by the American Society of Emergency Radiology (ASER) was incorporated to determine if such an approach might facilitate the entry, maintenance, and cataloguing of interesting cases. The RSNA MIRC software was obtained from the main MIRC website and installed. A coding system was developed based on the outline form of the ASER curriculum. Weekly reports were generated tallying the number of cases in each category of the curriculum. Resident participation in the entry and maintenance of cases markedly increased after incorporation of the ASER curriculum. The coding schema facilitated progress assessment. Ultimately, 454 total cases were entered into the MIRC database, representing at least 42% of the subcategories within the ASER curriculum (161 out of 376). The incorporation of the ASER emergency radiology curriculum greatly facilitated the location, cataloguing, tracking, and maintenance of representative cases and served as an effective means by which to unify the efforts of the department to develop a comprehensive teaching resource within this subspecialty. This approach and format will be extended to other educational curricula in other radiological subspecialties.     

Comparative cytogenetic study of spindle cell and pleomorphic leiomyosarcomas of soft tissues: a report from the CHAMP Study Group

Leiomyosarcomas (LMS) of soft tissues frequently show complex karyotypic changes, and no specific aberration has been identified. The aim of this study was to search for recurrent chromosome aberrations in soft tissue LMSs and to correlate these, if present, with morphological and clinical parameters. From a series of soft tissue sarcomas thoroughly reexamined cytogenetically and histopathologically, 45 LMSs were retrieved; 35 were classified microscopically as spindle cell, 3 as epithelioid, and 7 as pleomorphic. Clonal chromosome changes were present in 14, 3, and 3 cases, respectively. This series was combined with 11 previously published, karyotypically abnormal pleomorphic LMSs for cytogenetic-clinico-histopathological correlations. The breakpoints were widely scattered, with no predilection of any of the recurrent breakpoints and losses to any of the morphologic subtypes. Combining numerical and unbalanced structural changes, the most frequently lost segments were 3p21-p23 (11 cases), 8p21-pter, 13q12-q13, 13q32-qter (10 cases each), 1q42-qter, 2p15-pter, 18p11 (9 cases each), 1p36, 11q23-qter (8 cases each), and 10q23-qter (7 cases). The most frequent gain was 1q12-q31 (6 cases). There was a greater frequency of losses in 1p and 8p and a lower frequency of losses in 10q and 13q in tumors that had metastasized than in localized tumors. We conclude that LMSs with clonal abnormalities display highly complex karyotypic changes and extensive heterogeneity. No significant correlation exists between these changes and age and sex of the patients, or with depth of tumor, topography, microscopic subtype, or tumor grade. Losses in 1p36 and 8p21-pter may be associated with increased risk of metastases. Comparison of our findings in soft tissue LMS with those previously reported in LMS in other locations suggest that the karyotypic profile is more dependent on site of origin than on microscopic features.     

A survey of medical students on the impact of a new digital imaging library in the dissection room

Radiology has a recognised role in undergraduate anatomy education. The recent digitalisation of radiology has created new learning opportunities involving techniques such as image labelling, 3D reconstruction, and multiplanar reformatting. An opportunity was identified at the University of Nottingham to create a digital library of normal radiology images as a learner‐driven adjunct in anatomy dissection sessions. We describe the process of creating a de novo digital library by sourcing images for presentation at computer workstations. Students' attitudes towards this new resource were assessed using a questionnaire which used a 5 point Likert scale and also offered free text responses. One hundred and forty‐one out of 260 students (54%) completed the questionnaire. The most notable findings were: a positive response to the relevance of imaging to the session topics (median score 4), strong agreement that images should be available on the university website (median score 5), and disagreement that enough workstations were available (median score 2). About 24% of respondents suggested independently that images needed more labeling to help with orientation and identification. This first phase of supplying a comprehensive imaging library can be regarded as a success. Increasing availability and incorporating dynamic labeling are well recognized as important design concepts for electronic learning resources and these will be improved in the second phase of delivery as a direct result of studentfeedback. Hopefully other centers can benefit from this experience and will consider such a venture to be worthwhile. Clin. Anat. 22:761–769, 2009. © 2009 Wiley‐Liss, Inc.     

An Integrated Approach to a Teaching File Linked to PACS

To meet the educational needs of a medical imaging department with a strong teaching commitment, a teaching file that uses digital data supplied by the institutional picture archiving and communications system (PACS) was required. This teaching file had to be easily used by the end users, have a simple submission process, be able to support multiple users, be searchable on all data fields, and implementing the teaching file must not incur any additional software or hardware costs. The teaching file developed to address this problem takes advantage of the database structure and capabilities of several components included in the commercial PACS installed at the hospital. MS Access is used to seamlessly integrate with the digital imaging and communication in medicine (DICOM) database of a normal work station that is part of the PACS. This integration allows relevant patient and study demographics to be copied from images of interest and then to be stored in a separate database as the back-end of the digital teaching file. When images for a particular teaching file case need to be reviewed, they are automatically retrieved and displayed from the main PACS database using an open application programming interface (API) connection defined on the PACS web server. Utilizing this open API connection means the teaching file contains only the relevant demographic information of each teaching file case; no image data is stored locally. The open API connection allows access to imaging data usually not encountered in a teaching file, allowing more comprehensive imaging case files to be developed by the radiologist. Other advantages of this teaching file design are that it does not duplicate image data, it is small allowing simple ongoing backup, and it can be opened with multiple users accessing the database without compromising data access or integrity.     

INDIAM—An e-Learning System for the Interpretation of Mammograms

We propose the design of a teaching system named Interpretation and Diagnosis of Mammograms (INDIAM) for training students in the interpretation of mammograms and diagnosis of breast cancer. The proposed system integrates an illustrated tutorial on radiology of the breast, that is, mammography, which uses education techniques to guide the user (doctors, students, or researchers) through various concepts related to the diagnosis of breast cancer. The user can obtain informative text about specific subjects, access a library of bibliographic references, and retrieve cases from a mammographic database that are similar to a query case on hand. The information of each case stored in the mammographic database includes the radiological findings, the clinical history, the lifestyle of the patient, and complementary exams. The breast cancer tutorial is linked to a module that simulates the analysis and diagnosis of a mammogram. The tutorial incorporates tools for helping the user to evaluate his or her knowledge about a specific subject by using the education system or by simulating a diagnosis with appropriate feedback in case of error. The system also makes available digital image processing tools that allow the user to draw the contour of a lesion, the contour of the breast, or identify a cluster of calcifications in a given mammogram. The contours provided by the user are submitted to the system for evaluation. The teaching system is integrated with AMDI—An Indexed Atlas of Digital Mammograms—that includes case studies, e-learning, and research systems. All the resources are accessible via the Web.     

Mechanisms of inactivation of the p16INK4a gene in leiomyosarcoma of soft tissue: decreased p16 expression correlates with promoter methylation and poor prognosis

The p16INK4a tumour suppressor gene, encoding p16 protein, plays a crucial role in regulation of the G1 cell-cycle phase. To investigate the potential role of p16 in soft tissue leiomyosarcoma (LMS), an immunohistochemical analysis was performed of 77 LMSs for p16 expression. Decreased expression of the p16 protein was identified in 25 of 77 LMSs (32%). Decreased expression of p16 correlated significantly with large tumour size (p=0.0038). In a univariate analysis, large tumour size and decreased expression of p16 were statistically significant adverse prognostic factors (p=0.025 and p=0.0021, respectively). In a multivariate analysis including conventional clinicopathological parameters, decreased expression of p16 protein was revealed as the only independent unfavourable prognostic factor (p=0.012). To elucidate the mechanisms of inactivation of the p16INK4a gene, 49 LMSs for which genomic DNA was available were examined; analysis for homozygous deletion, mutation, and promoter hypermethylation was conducted using differential PCR, PCR-SSCP, and methylation-specific PCR, respectively. Promoter hypermethylation was detected in 11 of 49 LMS cases (22%); homozygous deletion was detected in 3 of 49 cases (6%); and mutation was not recognized in any of the cases studied. Eight of 15 cases (53%) with decreased expression of p16 protein revealed methylation of the p16INK4a gene promoter. Promoter hypermethylation correlated closely with decreased expression and poor prognosis (p=0.0014 and p=0.0088, respectively). These results suggest that decreased expression of p16 protein can be considered as an independent reliable prognostic parameter in patients with soft tissue LMS. Furthermore, promoter methylation was more frequent than either homozygous deletion or mutation in this tumour, and promoter methylation was also shown to have a strong association with inactivation of the p16INK4a gene.     

HEAL: an instructional design model applied to an online clerkship in family medicine.

OBJECTIVE: The potential of distance learning technology to deliver educational programs in which instruction and evaluation are of a consistent and high standard across multiple settings is hampered by a lack of instructional design models. In response, we developed the HEAL (Heuristic for Electronic Asynchronous Learning) model for designing online curricula. DESCRIPTION: HEAL is based on the theories that learning is facilitated by independent problem solving, investigation, and discovery (heuristics); collaboration between students fosters learning; and the proven educational cycle of practice, feedback, and reflection is integral to the interrelated domains of skill development and personal awareness.(1) The HEAL model is defined by synergistic online learning activities integrated with real patient care. It is applicable to all medical education levels. We applied this innovative design template to an online curriculum that augments our conventional six-week third-year clerkship. Our students, who were placed in distant family physician offices, needed more interaction and learning from peers and faculty. The three elements of HEAL, and implementation in the "online clerkship," are: (1) Didactic modules teach and illustrate concepts. Students study modules (HTML pages) on management of diabetes (DM), and complete five modules on evidence-based medicine (EBM). They do EBM literature searches reviewed online by peers, faculty, and librarians, who provide feedback. (2) A problem-based case discussion promotes application of concepts from modules (horizontal curricular integration). Students view streamed video of a patient with a history suggestive of diabetes, review her medical chart online, and suggest evidence-based management in an asynchronous discussion group. The case progresses weekly to mimic 12 months of continuity of care. (3) A collaborative journal activity explores the results of applying elements one and two to real patients (vertical integration). Additional elements advance reflection, professionalism, and medical humanism. Participation in the journal discussion group, stimulated by online readings, enhances self-awareness, informs psychosocial aspects of element 2, and promotes generalization of learning objectives to real patients. We use BlackBoard software. Students log in two to three times per week. Faculty who are trained in online moderation facilitate the threaded discussion groups and provide feedback. DISCUSSION: Students in alternating clerkship blocks complete the online clerkship. Their performance is compared with that of students who complete a face-to-face diabetes curriculum, but no curriculum on EBM or medical humanism. After nearly a year (105 students), compared with the non-online group, students completing the online clerkship demonstrated greater gains in reported EBM skills from preto post-clerkship, larger increases in mean score (from pre to post) on a medical-humanism aptitude scale, and higher scores on a post-clerkship diabetes management assessment (all comparisons p <.05). The online clerkship will become a permanent part of our clerkship and we have begun to use HEAL to design other online courses, including continuing education courses.     

Death-associated protein kinase (DAP kinase) alteration in soft tissue leiomyosarcoma: Promoter methylation or homozygous deletion is associated with a loss of DAP kinase expression

The death-associated protein kinase (DAP kinase) was initially identified as a positive mediator of programmed cell death induced by interferon-gamma. To investigate the potential role and the alteration of the DAP kinase gene in soft tissue leiomyosarcoma (LMS), we first searched for homozygous deletion and promoter hypermethylation in 45 LMSs for which genomic DNA was available, using differential PCR and methylation-specific PCR, respectively. Promoter methylation was recognized in 10 of 45 cases (22%), and homozygous deletion was detected in 3 of 45 cases (7%). p53 mutation was detected in 11 of 45 LMS cases (24%). Cases with DAP kinase alteration or p53 mutation showed a close correlation with high French Federation of Cancer Centers grade or with poor prognosis (P = 0.0244, P = 0.0491, respectively). Next, to determine that DAP kinase promoter methylation or homozygous deletion is involved in the down-regulation of DAP kinase expression, we examined the expression of DAP kinase protein by immunohistochemistry. Decreased expression of DAP kinase protein was recognized in 13 of 45 LMS cases (29%). Seven of 13 cases (54%) with decreased expression of DAP kinase protein revealed promoter methylation or homozygous deletion of DAP kinase, and the methylation status or homozygous deletion of its gene showed a close correlation with decreased DAP kinase expression (P = 0.0300). In conclusion, although DAP kinase alteration was relatively rare, DAP kinase alteration and/or p53 mutation may associate with tumor progression in soft-tissue LMSs. Furthermore, although further detailed analyses are necessary, promoter methylation or homozygous deletion status of DAP kinase may present a major alternative mechanism of a loss of or decrease in DAP kinase expression.     

The Importance of Human–Computer Interaction in Radiology E-learning

With the development of cross-sectional imaging techniques and transformation to digital reading of radiological imaging, e-learning might be a promising tool in undergraduate radiology education. In this systematic review of the literature, we evaluate the emergence of image interaction possibilities in radiology e-learning programs and evidence for effects of radiology e-learning on learning outcomes and perspectives of medical students and teachers. A systematic search in PubMed, EMBASE, Cochrane, ERIC, and PsycInfo was performed. Articles were screened by two authors and included when they concerned the evaluation of radiological e-learning tools for undergraduate medical students. Nineteen articles were included. Seven studies evaluated e-learning programs with image interaction possibilities. Students perceived e-learning with image interaction possibilities to be a useful addition to learning with hard copy images and to be effective for learning 3D anatomy. Both e-learning programs with and without image interaction possibilities were found to improve radiological knowledge and skills. In general, students found e-learning programs easy to use, rated image quality high, and found the difficulty level of the courses appropriate. Furthermore, they felt that their knowledge and understanding of radiology improved by using e-learning. In conclusion, the addition of radiology e-learning in undergraduate medical education can improve radiological knowledge and image interpretation skills. Differences between the effect of e-learning with and without image interpretation possibilities on learning outcomes are unknown and should be subject to future research.     

Dynamic “Inline” Images: Context-Sensitive Retrieval and Integration of Images into Web Documents

Integrating relevant images into web-based information resources adds value for research and education. This work sought to evaluate the feasibility of using “Web 2.0” technologies to dynamically retrieve and integrate pertinent images into a radiology web site. An online radiology reference of 1,178 textual web documents was selected as the set of target documents. The ARRS GoldMiner™ image search engine, which incorporated 176,386 images from 228 peer-reviewed journals, retrieved images on demand and integrated them into the documents. At least one image was retrieved in real-time for display as an “inline” image gallery for 87% of the web documents. Each thumbnail image was linked to the full-size image at its original web site. Review of 20 randomly selected Collaborative Hypertext of Radiology documents found that 69 of 72 displayed images (96%) were relevant to the target document. Users could click on the “More” link to search the image collection more comprehensively and, from there, link to the full text of the article. A gallery of relevant radiology images can be inserted easily into web pages on any web server. Indexing by concepts and keywords allows context-aware image retrieval, and searching by document title and subject metadata yields excellent results. These techniques allow web developers to incorporate easily a context-sensitive image gallery into their documents.     

Automatic Retrieval of Bone Fracture Knowledge Using Natural Language Processing

Natural language processing (NLP) techniques to extract data from unstructured text into formal computer representations are valuable for creating robust, scalable methods to mine data in medical documents and radiology reports. As voice recognition (VR) becomes more prevalent in radiology practice, there is opportunity for implementing NLP in real time for decision-support applications such as context-aware information retrieval. For example, as the radiologist dictates a report, an NLP algorithm can extract concepts from the text and retrieve relevant classification or diagnosis criteria or calculate disease probability. NLP can work in parallel with VR to potentially facilitate evidence-based reporting (for example, automatically retrieving the Bosniak classification when the radiologist describes a kidney cyst). For these reasons, we developed and validated an NLP system which extracts fracture and anatomy concepts from unstructured text and retrieves relevant bone fracture knowledge. We implement our NLP in an HTML5 web application to demonstrate a proof-of-concept feedback NLP system which retrieves bone fracture knowledge in real time.