Search

In brain tumor surgery, soft-tissue deformation, known as brain shift, introduces inaccuracies in the application of the preoperative surgical plan and impedes the advancement of image-guided surgical (IGS) systems. Considerable progress in using patient-specific biomechanical models to update the preoperative images intraoperatively has been made. These model-update methods rely on accurate intraoperative 3D brain surface displacements. In this work, we investigate and develop a fully automatic method to compute these 3D displacements for lengthy (~15 minutes) stereo-pair video sequences acquired during neurosurgery. The first part of the method finds homologous points temporally in the video and the second part computes the nonrigid transformation between these homologous points. Our results, based on parts of 2 clinical cases, show that this speedy and promising method can robustly provide 3D brain surface measurements for use with model-based updating frameworks.

Identification of the anterior and posterior commissure is crucial in stereotactic and functional neurosurgery, human brain mapping, and medical image processing. We present a learning-based algorithm to automatically and rapidly localize these landmarks using random forests regression. Given a point in the image, we extract a set of multi-scale long-range textural features, and associate a probability for this point to be the landmark. We build random forests models to learn the relationship between the value of these features and the probability of a point to be a landmark point. Three-stage coarse-to-fine models are trained for AC and PC separately using down-sampled by 4, down-sampled by 2, and the original images. Testing is performed in a hierarchical approach to first obtain a rough estimation at the coarse level and then fine-tune the landmark position. We extensively evaluate our method in a leave-one-out fashion using a large dataset of 100 T1-weighted images. We also compare our method to the state-of-art AC/PC detection methods including an atlas-based approach with six well-established nonrigid registration algorithms and a publicly available implementation of a model-based approach. Our method results in an overall error of 0.84±0.41mm for AC, 0.83±0.36mm for PC and a maximum error of 2.04mm; it performs significantly better than the model-based AC/PC detection method we compare it to and better than three of the nonrigid registration methods. It is much faster than nonrigid registration methods.