WebMILL Labeling System

URL: https://brassie.ece.jhu.edu/

Technologies used: Apache Wicket 1.3.5, Apache Spring, JPA, Hibernate EntityManager, JavaMail, C3P0, JBossCache2, Jersey 1.1, JAXB, Log4J,Tomcat 6.x, MySQL, XHTML, CSS 2.0, BlueprintCSS, JavaScript.
Java2D, Swing, Synth pluggable look and feel, JAXB (Java architecture for XML binding), HTTP client

WebMILL (Web-Based Tool for Medical Image Labeling Language) is a project conveyed jointly with John Hopkins University to design and develop a web interface for manual delineation of medical images.

The project covered creation of a web application, delineation tool applet and client-server communication layer. The web site, which functionality covers training users, providing feedback, and allowing users to monitor their progress. The site design was developed by Pragmasoft design studio, our pros covered slicing the design into valid cross-browser XHTML/CSS. BlueprintCSS technology was used for the slicing.

A delineation tool was developed on the basis on ImageJ, an open-source Java application. It appears in a pop-up browser window, with a specific task pre-loaded to it with an MRI image, where a user should delineate a specified target region.

Delineation tool has the following functions: paint/polygon/erase mode, brush size/brush color/brush transparency controls, brightness/contrast, picture zoom, reset painting to beginning. The user can skip the current task, switch between finish/continue modes, view an MRI slice above/below the current one.

With the help of this application, statistical atlases of different parts of the human body can be created, and they have proven to be extremely useful across a wide range of applications including those involving medical, scientific, and technological aims. They can be used to establish anatomical norms from which deviations can be used to diagnose or stage disease. They can also be used to establish modes of deviation within both normal and diseased populations from which populations can be classified according to numerous variates including age, sex, and disease.