Indexing of fetal anatomies from 3D ultrasound images

Research Area: A.4-Medical Imaging
Status: Finished  
Project leaders:
  • Gustavo Carneiro
Collaborators:
  • Fernando Amat
  • Bogdan Georgescu
  • Dorin Comaniciu
Proposed start date: 2007-06-08 Proposed end date: 2007-09-21
Description:

The use of 3D ultrasound data has several advantages over 2D ultrasound for fetal biometric measurements, such as considerable decrease in the examination time, possibility of post-exam data processing by experts and the ability to produce 2D views of the fetal anatomies in orienta- tions that cannot be seen in common 2D ultrasound exams. However, the search for standardized planes and the precise localization of fetal anatomies in ultrasound volumes are hard and time consuming processes even for expert physicians and sonographers. The relative low resolution in ultrasound volumes, small size of fetus anatomies and inter- volume position, orientation and size variability make this localization problem even more challenging. In order to make the plane search and fetal anatomy localization prob- lems completely automatic, we introduce a novel principled probabilistic model that combines discriminative and generative classifiers with contextual information and sequential sampling. We implement a system based on this model, where the user queries consist of semantic keywords that represent anatomical structures of interest. After queried, the system automatically displays standardized planes and produces biometric measurements of the fetal anatomies. Experimental results on a held-out test set show that the automatic measurements are within the interuser variability of expert users. It resolves for position, orientation and size of three different anatomies in less than 10 seconds in a dual-core computer running at 1.7 GHz (Fig. 1).

 

Figure 1: Different anatomies automatically detected in an unseen 3D ultrasound image using the approach developed in this project. The algorithm detects in less than 10 secs. position, orientation and size each anaotmy of interest. (From [1]).

-----------------------------------------------------------------------------

[1] G. Carneiro, F. Amat, B. Georgescu, S. Good, and D. Comaniciu, “Semantic-based indexing of fetal anatomies from 3-D ultrasound data using global/semi-local context and sequential sampling,” in 2008 CVPR, Anchorage, USA, June 2008, pp. 1-8.