Creation of a Database-Driven Website for QuickTime Streaming Neurological Teaching Videos

Document Type

Conference Proceeding

Publication Date

2001

Abstract

The use of patients and video recordings to teach neurology in the classroom is a valuable method of integrating basic neuroscience principles with physical diagnosis. However, this teaching method is not without practical limitations. Patient availability is often uncertain and unreliable. Analog videotape is costly to duplicate and distribute for unlimited review by individual students. And, neither format provides a convenient way for individual students to review and reinforce parts of the physical exam that may be particularly helpful to their learning. Our long range objectives are to create digital source videos depicting neurological disorders, prepare them for low bandwidth Internet distribution, and organize them into a searchable database on an Internet server from which they may be distributed, free of charge, to health care educational institutions.

Our intent is to have client institutions distribute these videos 'on-demand' via Local Area Network to students within their respective institutions. We feel this will help remove existing limitations to the use of live patients and video as an educational tool, and help bring the testing of performance-based knowledge to a new level in health care education. We previously described capture and compression parameters used to prepare neurologic QuickTime videos for streaming delivery (Pearson, et. al., 1999, SOL/CHES Proceedings, Philadelphia).

The present report describes the materials and strategies used to establish a website at which these streaming videos are contained in a database searchable through the HTML environment. Our site (www.ntv.wright.edu) is contained on a Macintosh G3 computer (400mHz; 256MB RAM) running Mac OS X-Apache software, and maintained in the Wright State University School of Medicine Department of Anatomy. In designing this site, our objectives were to: (1) allow onsite searches, browsing and preview of video files; (2) allow external searches to recognize all video files, especially clips of individual signs; and (3) maintain password protection for patient privacy.

We achieved these objectives through creation of a database-driven website built around: (1) FileMaker 5 Developer database software, (2) Dreamweaver web building software, and (3) Lasso software designed to allow the FileMaker database to be accessed through an HTML environment. Video files at our site consist of either:

  1. short 'clips' that depict single neurologic signs, or symptoms
  2. and (2)'full length movies' that incorporate short 'clips' of a

particular patient into one continuous file. Of these two file types, only the short 'clips' are viewable over open Internet directly from our site. Consequently, the short 'clip' files are placed within the FileMaker records and used as searchable 'preview' files. FileMaker records are indexed using keywords consistent with Medical Subject Heading (MeSH) vocabulary. They also contain fields that identify relevant patient diagnostic information and point to "Patient Profile" pages (HTML pages) that contain all downloadable files of that patient (i.e., 'full length movie' files; neuroimaging picture files; case history text files, etc.). Lasso Web Data Engine software allows the FileMaker database program to be accessed through the HTML environment. A search and browse interface allows access to database records as well as the ëPatient Profileí pages. The search and browse page is preceded by a password acquisition page. 'Dummyí HTML pages containing MeSH keywords in their meta tag regions provide exposure to external searches and point to the front door of our site. This strategy makes files accessible to onsite searching and browsing, as well as to external search programs, such as those designed to provide decision support in the primary care clinical environment.

Comments

Presented at the 2001 Slice of Life Workshop, Munich, Germany.

Link to Full Text


Share

COinS