2013 Summer Project Week:Sobolev Segmenter

• 

  Projects List

• 

  The 2D segmentation of tumor by Sobolev active contour.

• 

  The 3D segmentation of tumor by Sobolev active contour.

Key Investigators

• UAB: Arie Nakhmani, LiangJia Zhu, Allen Tannenbaum
• BWH: Yi Gao, Ron Kikinis
• Utah: Rob MacLeod, Josh Cates

Objective

General segmentation tools can be used in a wide range of biomedical applications from tumor delineation to segmentation of the atrial wall used in the research on the atrial fibrillation problem. Sobolev active contour is a general 2D segmenter that can be used in the applications above. It can be extended for medical volume segmentation. Our objective is to implement Slicer's extension based on Sobolev active contours algorithm for volume segmentation.

Approach, Plan

We prepare C++ implementation of Sobolev active contour algorithm, and convert it to Slicer Commandline extension. This version needs a few sparse initial contours on some slices of the segmented volume.

In another approach, we implement similar algorithm in Python, as an Editor effect of Slicer. In this case, only the definition of 3D depth, and a single click inside the area of the segmented target are needed for the algorithm to work.

Progress

Alpha versions of both approaches (C++ Commandline and Python Editor effect) have been prepared.

Delivery Mechanism

This work will be delivered to the NA-MIC Kit as a loadable Commandline extension and an Editor effect.

References

• A. Nakhmani, A. Tannenbaum, "Tracking with Adaptive Sobolev Snakes." Submitted to

IEEE Transactions on Image Processing.

• A. Nakhmani, A. Tannenbaum, "Self-Crossing Detection and Location for Parametric

Active Contours," IEEE Transactions on Image Processing, DOI:10.1109/TIP.2012.2188808, Volume 21, Issue 7, pp. 3150-3156, July 2012.