3D Intervertebral disc localization and segmentation from MR images by data-driven regression and classification

In this paper we propose a new fully-automatic method for localizing and segmenting 3D intervertebral discs from MR images, where the two problems are solved in a unified data-driven regression and classification framework. We estimate the output (image displacements for localization, or fg/bg labels for segmentation) of image points by exploiting both training data and geometric constraints simultaneously. The problem is formulated in a unified objective function which is then solved globally and efficiently. We validate our method on MR images of 25 patients. Taking manually labeled data as the ground truth, our method achieves a mean localization error of 1.3 mm, a mean Dice metric of 87%, and a mean surface distance of 1.3 mm. Our method can be applied to other localization and segmentation tasks.

Enterprise localization practices and their implication for knowledge management in foreign participation in the Chinese construction sector

Foreign firms active in the Chinese construction industry find themselves in a competitive environment unlike the environments in which they operate back home. Adaptations are therefore required in order that firms retain competitive leverage while integrating into the circumstances of the Chinese market. Since these firms primarily compete on superior knowledge capabilities it is instructive to understand how adaptations both serve to transfer propriety capabilities to China while adapting to the knowledge characteristics of the Chinese market into which those capabilities are transplanted. Five localizing parameters across which adaptations take place are identified in this paper and rates of localization are tabulated for 60 foreign firms active in the Chinese construction industry. Revealed localizing behaviors are then analyzed in the light of existing literature for explanatory power. It is concluded that in order to be strategically competitive firms must not only retain knowledge differentials, but that the knowledge must be coded and disseminated in a manner suited to the environment in which it is to be utilized.

The effect of deformation twinning on stress localization in a three dimensional TWIP steel microstructure

We present an investigation of the effect of deformation twinning on the visco-plastic response and stress localization in a low stacking fault energy twinning-induced plasticity (TWIP) steel under uniaxial tension loading. The three-dimensional full field response was simulated using the fast Fourier transform method. The initial microstructure was obtained from a three dimensional serial sectionusing electron backscatter diffraction. Twin volume fraction evolution upon strain was measured so the hardening parameters of the simple Voce model could be identified to fit both the stress-strain behavior and twinning activity. General trends of texture evolution were acceptably predicted including the typical sharpening and balance between the 1 1 1 fiber and the 1 0 0 fiber. Twinning was found to nucleate preferentially at grain boundaries although the predominant twin reorientation scheme did not allow spatial propagation to be captured. Hot spots in stress correlated with the boundaries of twinned voxel domains, which either impeded or enhanced twinning based on which deformation modes were active locally.

Localization and degmentation of 3D intervertebral discs in MR Images by data driven estimation

This paper addresses the problem of fully-automatic localization and segmentation of 3D intervertebral discs (IVDs) from MR images. Our method contains two steps, where we first localize the center of each IVD, and then segment IVDs by classifying image pixels around each disc center as foreground (disc) or background. The disc localization is done by estimating the image displacements from a set of randomly sampled 3D image patches to the disc center. The image displacements are estimated by jointly optimizing the training and test displacement values in a data-driven way, where we take into consideration both the training data and the geometric constraint on the test image. After the disc centers are localized, we segment the discs by classifying image pixels around disc centers as background or foreground. The classification is done in a similar data-driven approach as we used for localization, but in this segmentation case we are aiming to estimate the foreground/background probability of each pixel instead of the image displacements. In addition, an extra neighborhood smooth constraint is introduced to enforce the local smoothness of the label field. Our method is validated on 3D T2-weighted turbo spin echo MR images of 35 patients from two different studies. Experiments show that compared to state of the art, our method achieves better or comparable results. Specifically, we achieve for localization a mean error of 1.6-2.0 mm, and for segmentation a mean Dice metric of 85%-88% and a mean surface distance of 1.3-1.4 mm.

Fully automatic localization and segmentation of 3D vertebral bodies from CT/MR images via a learning-based method

In this paper, we address the problems of fully automatic localization and segmentation of 3D vertebral bodies from CT/MR images. We propose a learning-based, unified random forest regression and classification framework to tackle these two problems. More specifically, in the first stage, the localization of 3D vertebral bodies is solved with random forest regression where we aggregate the votes from a set of randomly sampled image patches to get a probability map of the center of a target vertebral body in a given image. The resultant probability map is then further regularized by Hidden Markov Model (HMM) to eliminate potential ambiguity caused by the neighboring vertebral bodies. The output from the first stage allows us to define a region of interest (ROI) for the segmentation step, where we use random forest classification to estimate the likelihood of a voxel in the ROI being foreground or background. The estimated likelihood is combined with the prior probability, which is learned from a set of training data, to get the posterior probability of the voxel. The segmentation of the target vertebral body is then done by a binary thresholding of the estimated probability. We evaluated the present approach on two openly available datasets: 1) 3D T2-weighted spine MR images from 23 patients and 2) 3D spine CT images from 10 patients. Taking manual segmentation as the ground truth (each MR image contains at least 7 vertebral bodies from T11 to L5 and each CT image contains 5 vertebral bodies from L1 to L5), we evaluated the present approach with leave-one-out experiments. Specifically, for the T2-weighted MR images, we achieved for localization a mean error of 1.6 mm, and for segmentation a mean Dice metric of 88.7% and a mean surface distance of 1.5 mm, respectively. For the CT images we achieved for localization a mean error of 1.9 mm, and for segmentation a mean Dice metric of 91.0% and a mean surface distance of 0.9 mm, respectively.

A hybrid localization approach in 3D wireless sensor network

Location information acquisition is crucial for many wireless sensor network (WSN) applications. While existing localization approaches mainly focus on 2D plane, the emerging 3D localization brings WSNs closer to reality with much enhanced accuracy. Two types of 3D localization algorithms are mainly used in localization application: the range-based localization and the range-free localization. The range-based localization algorithm has strict requirements on hardware and therefore is costly to implement in practice. The range-free localization algorithm reduces the hardware cost but at the expense of low localization accuracy. On addressing the shortage of both algorithms, in this paper, we develop a novel hybrid localization scheme, which utilizes the range-based attribute RSSI and the range-free attribute hopsize, to achieve accurate yet low-cost 3D localization. As anchor node deployment strategy plays an important role in improving the localization accuracy, an anchor node configuration scheme is also developed in this work by utilizing the MIS (maximal independent set) of a network. With proper anchor node configuration and propagation model selection, using simulations, we show that our proposed algorithm improves the localization accuracy by 38.9% compared with 3D DV-HOP and 52.7% compared with 3D centroid.

RSSI-based localization of a wireless sensor node with a flying robot

We consider the problem of navigating a ying robot to a specific sensor node within a wireless sensor network. This target sensor node periodically sends out beacons. The robot is capable of sensing the received signal strength of each received beacon (RSSI measurements). Existing approaches for solving the sensor spotting problem with RSSI measurements do not deal with noisy channel conditions and/or heavily depend on additional hardware capabilities. In this work we reduce RSSI uctuations due to noise by continuously sampling RSSI values and maintaining an exponential moving average (EMA). The EMA values enable us to detect significant decrease of the received signal strength. In this case it is reasoned that the robot is moving away from the sensor. We present two basic variants to decide a new moving direction when the robot moves away from the sensor. Our simulations show that our approaches outperform competing algorithms in terms of success rate and ight time. Infield experiments with real hardware, a ying robocopter successfully and quickly landed near a sensor placed in an outdoor test environment. Traces show robustness to additional environmental factors not accounted for in our simulations.

Evaluation and comparison of 3D intervertebral disc localization and segmentation methods for 3D T2 MR data: A grand challenge.

The evaluation of changes in Intervertebral Discs (IVDs) with 3D Magnetic Resonance (MR) Imaging (MRI) can be of interest for many clinical applications. This paper presents the evaluation of both IVD localization and IVD segmentation methods submitted to the Automatic 3D MRI IVD Localization and Segmentation challenge, held at the 2015 International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI2015) with an on-site competition. With the construction of a manually annotated reference data set composed of 25 3D T2-weighted MR images acquired from two different studies and the establishment of a standard validation framework, quantitative evaluation was performed to compare the results of methods submitted to the challenge. Experimental results show that overall the best localization method achieves a mean localization distance of 0.8 mm and the best segmentation method achieves a mean Dice of 91.8%, a mean average absolute distance of 1.1 mm and a mean Hausdorff distance of 4.3 mm, respectively. The strengths and drawbacks of each method are discussed, which provides insights into the performance of different IVD localization and segmentation methods.