Rasika Rajapakshe PhD, Presenter: Nothing to Disclose

Steve McAvoy BSC, Abstract Co-Author: Nothing to Disclose

Christabelle Bitgood BSC, Abstract Co-Author: Nothing to Disclose

Elena Ostroumov PhD, Abstract Co-Author: Nothing to Disclose

We have developed a software framework that can successfully store and retrieve respondent data for Breast Cancer Risk Assessment research survey.

A software framework from which future researchers may easily deploy surveys online, analyze results in real time, and export their data in open formats for Breast Cancer Risk Assessment (BCRA) was needed to estimate the Breast Cancer Risk distribution in the province of British Columbia.

A sample of surveys were analysed and broken down into their common elements. Most BCRA survey questions require a respondent to select one or multiple answers from a list of pre-defined choices. These types of questions are easily generalized into a simple, yet flexible, database schema and easily reproduced online using HTML, a multi-user relational database, an HTTP server, and the PHP programming language. BCRA survey questions are categorized into one of two types: pre-defined response or user-supplied response. Questions within the pre-defined response category are further broken down to determine if one or multiple responses are required. Questions categorized as user-supplied are divided into two new subtypes: integer-based or text-based. The questions and associated choices are then entered into an SQL-compliant relational database. Using the framework’s API, websites can extract a list of questions and display them as desired over one or many web pages. HTML output can be easily customized using standard CSS rules and a JavaScript validation library complements the framework to enforce any restrictions on user input, such as mandatory input fields. Respondents’ BCRA survey data is stored in a foreign key relationship allowing for an accurate interpretation of results.

The framework has been successfully implemented within the Breast Cancer Risk Assessment Project (http://www.breastcancerrisk.ca) which has been collecting data from respondents on a daily basis since September of 2009. The framework has recorded over 160,000 responses to various questions from more than 3,200 unique respondents. Using the API, data has been successfully exported to Microsoft Excel and other third party analysis software.

Rajapakshe, R, McAvoy, S, Bitgood, C, Ostroumov, E, A Web-based Survey Software Framework for Rapid Survey Deployment and Results Analysis for Breast Cancer Risk Assessment.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11010228.html