|
|
The Evaluation of Automated Boundary Detection Algorithms in Ultrasound ImagesIt is a goal of our image processing group to develop a reliable technique for the automatic identification of the boundary of body organ or tissue type in ultrasound images. However, the automated analysis of ultrasound images is fraught with difficulty because of several characteristics of the images, notably: poor spatial resolution; noise due to speckle; and limited soft tissue contrast. Any automatic boundary detection method must be evaluated against multiple expert observers’ hand-outlined boundaries, as the gold standard boundary for comparison is typically unavailable in medical images. Statistical evaluation of a boundary detection algorithm requires a large number of clinical data set, which often leads to a tedious and time-consuming data collection. Here we developed a web-based system to solve this problem, by getting the collection done via the Internet. The registered expert observers can login to the server from any places at any time, outline the medical images in a standard image database with the help of computerized outlining tools, then submit the outlines on line. An evaluation program follows up to evaluate computer outlining algorithms by comparing their outlines with those experts’ outlines. We have posted a group of transrectal ultrasound images of the prostate and a group of breast ultrasound images in the image database.
Related links :Instructions to use the web tools related with the CIMIL images archive Consult the CIMIL Images Archive Upload Images in the CIMIL Archive
Publications related to Assessment.
We would be glad if you could sign our guest book.
For more information, please contact the principal investigator: A/P Ng Wan Sing
|
