Germán González

Automated Reconstruction of Dendritic and Axonal Trees by Global Optimization with Geometric Priors

We present a novel probabilistic approach to fully automated delineation of tree structures in noisy 2D images and 3D image stacks. Unlike earlier methods that rely mostly on local evidence, ours builds a set of candidate trees over many different subsets of points likely to belong to the optimal tree and then chooses the best one according to a global objective function that combines image evidence with geometric priors. Since the best tree does not necessarily span all the points, the algorithm is able to eliminate false detections while retaining the correct tree topology. Manually annotated brightfield micrographs, retinal scans and the DIADEM challenge datasets are used to evaluate the performance of our method. We used the DIADEM metric to quantitatively evaluate the topological accuracy of the reconstructions and showed that the use of the geometric regularization yields a substantial improvement.

Steerable Features for Statistical 3D Dendrite Detection

Most state-of-the-art algorithms for filament detection in 3-D image-stacks rely on computing the Hessian matrix around individual pixels and labeling these pixels according to its eigenvalues. This approach, while very effective for clean data in which linear structures are nearly cylindrical, loses its effectiveness in the presence of noisy data and irregular structures. In this paper, we show that using steerable filters to create rotationally invariant features that include higher-order derivatives and training a classifier based on these features lets us handle such irregular structures. This can be done reliably and at acceptable computational cost and yields better results than state-of-the-art methods.

Delineating trees in noisy 2D images and 3D image-stacks

We present a novel approach to fully automated delineation of tree structures in noisy 2D images and 3D image stacks. Unlike earlier methods that rely mostly on local evidence, our method builds a set of candidate trees over many different subsets of points likely to belong to the final one and then chooses the best one according to a global objective function. Since we are not systematically trying to span all nodes, our algorithm is able to eliminate noise while retaining the right tree structure. Manually annotated dendrite micrographs and retinal scans are used to evaluate the performance of our method, which is shown to be able to reject noise while retaining the tree structure.

Tracking human motion and actions for interactive robots

A method is presented for kinematic pose estimation and action recognition from monocular robot vision through the use of dynamical human motion vocabularies. We propose the utilization of dynamical motion vocabularies towards bridging the decision making of observed humans and information from robot sensing. Our motion vocabulary is comprised of learned primitives that structure the action space for decision making and describe human movement dynamics. Given image observations over time, each primitive infers on pose independently using its prediction density on movement dynamics in the context of a particle filter. Pose estimates from a set of primitives inferencing in parallel are arbitrated to estimate the action being performed. The efficacy of our approach is demonstrated through tracking and action recognition over extended motion trials. Results evidence the robustness of the algorithm with respect to unsegmented multi-action movement, movement speed, and camera viewpoint.

Learning rotational features for filament detection

State-of-the-art approaches for detecting filament-like structures in noisy images rely on filters optimized for signals of a particular shape, such as an ideal edge or ridge. While these approaches are optimal when the image conforms to these ideal shapes, their performance quickly degrades on many types of real data where the image deviates from the ideal model, and when noise processes violate a Gaussian assumption. In this paper, we show that by learning rotational features, we can outperform state-of-the-art filament detection techniques on many different kinds of imagery. More specifically, we demonstrate superior performance for the detection of blood vessel in retinal scans, neurons in brightfield microscopy imagery, and streets in satellite imagery.

Photometric stereo endoscopy

Abstract. While color video endoscopy has enabled wide-field examination of the gastrointestinal tract, it often misses or incorrectly classifies lesions. Many of these missed lesions exhibit characteristic three-dimensional surface topographies. An endoscopic system that adds topographical measurements to conventional color imagery could therefore increase lesion detection and improve classification accuracy. We introduce photometric stereo endoscopy (PSE), a technique which allows high spatial frequency components of surface topography to be acquired simultaneously with conventional two-dimensional color imagery. We implement this technique in an endoscopic form factor and demonstrate that it can acquire the topography of small features with complex geometries and heterogeneous optical properties. PSE imaging of ex vivo human gastrointestinal tissue shows that surface topography measurements enable differentiation of abnormal shapes from surrounding normal tissue. Together, these results confirm that the topographical measurements can be obtained with relatively simple hardware in an endoscopic form factor, and suggest the potential of PSE to improve lesion detection and classification in gastrointestinal imaging.

Automated Delineation of Dendritic Networks in Noisy Image Stacks

We present a novel approach to 3D delineation of dendritic networks in noisy image stacks. We achieve a level of automation beyond that of state-of-the-art systems, which model dendrites as continuous tubular structures and postulate simple appearance models. Instead, we learn models from the data itself, which make them better suited to handle noise and deviations from expected appearance. From very little expert-labeled ground truth, we train both a classifier to recognize individual dendrite voxels and a density model to classify segments connecting pairs of points as dendrite-like or not. Given these models, we can then trace the dendritic trees of neurons automatically by enforcing the tree structure of the resulting graph. We will show that our approach performs better than traditional techniques on brighfield image stacks.

3D imaging techniques for improved colonoscopy

Colonoscopy screening with a conventional 2D colonoscope is known to reduce mortality due to colorectal cancer by half. Unfortunately, the protective value of this procedure is limited by missed lesions. To improve the sensitivity of colonoscopy to precancerous lesions, 3D imaging techniques could be used to highlight their characteristic morphology. While 3D imaging has proved beneficial for laparoscopic procedures, more research is needed to assess how it will improve applications of flexible endoscopy. In this editorial, we discuss the possible uses of 3D technologies in colonoscopy and factors that have hindered the translation of 3D imaging to flexible endoscopy. Emerging 3D imaging technologies for flexible endoscopy have the potential to improve sensitivity, lesion resection, training and automated lesion detection. To maximize the likelihood of clinical adoption, these technologies should require minimal hardware modification while maintaining the robustness and quality of regular 2D imaging.

Reconstructing geometrically consistent tree structures from noisy images

We present a novel approach to fully automated reconstruction of tree structures in noisy 2D images. Unlike in earlier approaches, we explicitly handle crossovers and bifurcation points, and impose geometric constraints while optimizing a global cost function. We use manually annotated retinal scans to evaluate our method and demonstrate that it brings about a very substantial improvement.

On the Relevance of Sparsity for Image Classification

In this paper we empirically analyze the importance of sparsifying representations for classification purposes. We focus on those obtained by convolving images with linear filters, which can be either hand designed or learned, and perform extensive experiments on two important Computer Vision problems, image categorization and pixel classification. To this end, we adopt a simple modular architecture that encompasses many recently proposed models. The key outcome of our investigations is that enforcing sparsity constraints on features extracted in a convolutional architecture does not improve classification performance, whereas it does so when redundancy is artificially introduced. This is very relevant for practical purposes, since it implies that the expensive run-time optimization required to sparsify the representation is not always justified, and therefore that computational costs can be drastically reduced.