Scalable Nearest Neighbor Algorithms for High Dimensional Data.

For many computer vision and machine learning problems, large training sets are key for good performance. However, the most computationally expensive part of many computer vision and machine learning algorithms consists of finding nearest neighbor matches to high dimensional vectors that represent the training data. We propose new algorithms for approximate nearest neighbor matching and evaluate and compare them with previous algorithms. For matching high dimensional features, we find two algorithms to be the most efficient: the randomized k-d forest and a new algorithm proposed in this paper, the priority search k-means tree. We also propose a new algorithm for matching binary features by searching multiple hierarchical clustering trees and show it outperforms methods typically used in the literature. We show that the optimal nearest neighbor algorithm and its parameters depend on the data set characteristics and describe an automated configuration procedure for finding the best algorithm to search a particular data set. In order to scale to very large data sets that would otherwise not fit in the memory of a single machine, we propose a distributed nearest neighbor matching framework that can be used with any of the algorithms described in the paper. All this research has been released as an open source library called fast library for approximate nearest neighbors (FLANN), which has been incorporated into OpenCV and is now one of the most popular libraries for nearest neighbor matching.

Tissue tracking and registration for image-guided surgery.

Vision-based tracking of tissue is a key component to enable augmented reality during a surgical operation. Conven- tional tracking techniques in computer vision rely on identifying strong edge features or distinctive textures in a well-lit environ- ment; however endoscopic tissue images do not have strong edge features, are poorly lit and exhibit a high degree of specular reflection. Therefore, prior work in achieving densely populated 3D features for describing tissue surface profiles require complex image processing techniques and have been limited in providing stable, long-term tracking or real-time processing. In this paper, we present an integrated framework for ac- curately tracking tissue in surgical stereo-cameras at real-time speeds. We use a combination of the STAR feature detector and Binary Robust Independent Elementary Features to acquire salient features that can be persistently tracked at high frame rates. The features are then used to acquire a densely-populated map of the deformations of tissue surface in 3D. We evaluate the method against popular feature algorithms in in-vivo animal study video sequences, and we also apply the proposed method to human partial nephrectomy video sequences. We extend the salient feature framework to support region tracking in order to maintain the spatial correspondence of a tracked region of tissue or a medical image registration to the surrounding tissue. In-vitro tissue studies show registration accuracies of 1.3-3.3 mm using a rigid-body transformation method.

Expression of vascular endothelial growth factor (VEGF)-C and VEGF-D, and their receptor VEGFR-3, during different stages of cervical carcinogenesis.

Cervical carcinogenesis has well-defined stages of disease progression including three grades of pre-invasive lesions--cervical intraepithelial neoplasia grades 1-3 (CIN 1-3)--and invasive cervical cancer. However, the biological properties of CIN lesions prone to develop invasive disease are not well defined. Recent observations suggest that early invasive disease spreads to regional lymph nodes in several tumour types and that growth factors (VEGF-C and VEGF-D) involved in new lymphatic vessel formation may play a crucial role in this process. The present study has assessed the expression of VEGF-C and VEGF-D, and their receptor VEGFR-3, in 152 cervical lesions (33 CIN 1, 33 CIN 2, 37 CIN 3, and 49 squamous cell carcinomas) to determine whether expression of lymphangiogenic factors occurs prior to invasion. The presence of lymphatic vessels was determined using LYVE-1 and podoplanin staining, as well as double immunostaining for LYVE-1/CD34 and podoplanin/CD34. In situ hybridization was performed to determine VEGFR-3 mRNA expression. A significant positive correlation was found between VEGF-C, VEGF-D, and VEGFR-3 expression through the different stages of cervical carcinogenesis. Significant differences in protein expression for VEGF-C, VEGF-D, and VEGFR-3 were found between CIN 1-2 and CIN 3 (p<0.001 for all), but not between CIN 3 and cervical cancer. More than 50% of the CIN 3 lesions showed moderate to strong staining for VEGF-C and VEGF-D, whereas most of the early pre-cancerous lesions (CIN 1 and 2) were negative. In cervical cancer, similar observations to those in CIN 3 were found. VEGFR-3 mRNA expression was found in the cytoplasm of epithelial neoplastic cells and VEGFR3 protein expression was found in more than 50% of CIN 3 lesions and cervical cancers, compared with 15% in CIN 1 and 2. These findings suggest an autocrine growth stimulation pattern via VEGFR-3. Adjacent CIN 3 was present in nine cervical cancers and displayed strong expression for VEGF-C, VEGF-D, and VEGFR-3. These results suggest that in cervical carcinogenesis a switch to the lymphangiogenic phenotype may occur at the stage of CIN 3.