Multi-layer Fusion Research Articles

Stereoscopic virtual realistic surgical simulation in intracranial aneurysms

Three-dimensional (3D)-computed tomographic angiography (CTA) has been widely used for surgical simulation of intracranial aneurysms. Stereo imaging technology is progressing rapidly in recent years and stereo imaging may make more realistic surgical simulation possible. Therefore, we aimed at the establishment of a technique for stereoscopic viewing of minute volume rendering images while pursuing a low cost. Between January 2009 and June 2011, 54 patients with ruptured intracranial aneurysms were enrolled in this study. CTA data was transferred to the workstation equipped with image-processing software, and multilayer fusion images were processed by neurosurgeons. Image data for stereoscopic viewing of multilayer fusion image from arbitrary directions were collected form rotational trajectories around an aneurysm and were output to MPEG file. Stereoscopic viewing using MPEG data was achieved by the freeware named Stereo Movie Maker. Stereo viewing method using QuickTime VR format was also tried. Multilayer fusion image created from CTA data displayed clearly the anatomical information about not only the aneurysm but also the surrounding structures, such as parent artery, venous system, brain tissue, skull bone, and scalp. The quality of the resulting multilayer fusion image was suitable for surgical simulation with virtual reality. Virtual realistic surgical simulation became possible by the combination of minute multilayer fusion image and stereoscopic viewing by our method. Our method for stereo viewing of multilayer fusion images resulted in an improvement in the capability of diagnostic imaging and the image-guided support for neurosurgical procedures in intracranial aneurysm.

Dynamic selection approaches for multiple classifier systems

In this paper we propose a new approach for dynamic selection of ensembles of classifiers. Based on the concept named multistage organizations, the main objective of which is to define a multi-layer fusion function adapted to each recognition problem, we propose dynamic multistage organization (DMO), which defines the best multistage structure for each test sample. By extending Dos Santos et al.’s approach, we propose two implementations for DMO, namely DSA m and DSA c . While the former considers a set of dynamic selection functions to generalize a DMO structure, the latter considers contextual information, represented by the output profiles computed from the validation dataset, to conduct this task. The experimental evaluation, considering both small and large datasets, demonstrated that DSA c dominated DSA m on most problems, showing that the use of contextual information can reach better performance than other existing methods. In addition, the performance of DSA c can also be enhanced in incremental learning. However, the most important observation, supported by additional experiments, is that dynamic selection is generally preferred over static approaches when the recognition problem presents a high level of uncertainty.

Multi-channel and multi-layer cell segmentations and decision fusion based on an hierarchical structure

In this paper, several multi-channel and multi-layer cell segmentation algorithms are developed and optimised individually. Then, a hierarchical structure is used in decision fusion to combine the four independent segmentation results to increase the probability of final cell detection. Experimental results have proved that both the proposed four independent edge detection and ‘CFSE’ based cell segmentation algorithms and the fusion algorithm are very robust. The statistical multiple frame experiments have shown that in 21 frames their average final detection rate is about 98%, with the 2% miss detection rate, and about 3% false detection rate.

Improved sequence-based prediction of disordered regions with multilayer fusion of multiple information sources

Motivation: Intrinsically disordered proteins play a crucial role in numerous regulatory processes. Their abundance and ubiquity combined with a relatively low quantity of their annotations motivate research toward the development of computational models that predict disordered regions from protein sequences. Although the prediction quality of these methods continues to rise, novel and improved predictors are urgently needed.Results: We propose a novel method, named MFDp (Multilayered Fusion-based Disorder predictor), that aims to improve over the current disorder predictors. MFDp is as an ensemble of 3 Support Vector Machines specialized for the prediction of short, long and generic disordered regions. It combines three complementary disorder predictors, sequence, sequence profiles, predicted secondary structure, solvent accessibility, backbone dihedral torsion angles, residue flexibility and B-factors. Our method utilizes a custom-designed set of features that are based on raw predictions and aggregated raw values and recognizes various types of disorder. The MFDp is compared at the residue level on two datasets against eight recent disorder predictors and top-performing methods from the most recent CASP8 experiment. In spite of using training chains with ≤25% similarity to the test sequences, our method consistently and significantly outperforms the other methods based on the MCC index. The MFDp outperforms modern disorder predictors for the binary disorder assignment and provides competitive real-valued predictions. The MFDp's outputs are also shown to outperform the other methods in the identification of proteins with long disordered regions.Availability: http://biomine.ece.ualberta.ca/MFDp.htmlSupplementary information: Supplementary data are available at Bioinformatics online.Contact: lkurgan@ece.ualberta.ca

Intelligent Biometric Information Management

We advance here a novel methodology for robust intelligent biometric information management with inferences and predictions made using randomness and complexity concepts. Intelligence refers to learning, adap- tation, and functionality, and robustness refers to the ability to handle incomplete and/or corrupt adversarial information, on one side, and image and or device variability, on the other side. The proposed methodology is model-free and non-parametric. It draws support from discriminative methods using likelihood ratios to link at the conceptual level biometrics and forensics. It further links, at the modeling and implementation level, the Bayesian framework, statistical learning theory (SLT) using transduction and semi-supervised lea- rning, and Information Theory (IY) using mutual information. The key concepts supporting the proposed methodology are a) local estimation to facilitate learning and prediction using both labeled and unlabeled data; b) similarity metrics using regularity of patterns, randomness deficiency, and Kolmogorov complexity (similar to MDL) using strangeness/typicality and ranking p-values; and c) the Cover – Hart theorem on the asymptotical performance of k-nearest neighbors approaching the optimal Bayes error. Several topics on biometric inference and prediction related to 1) multi-level and multi-layer data fusion including quality and multi-modal biometrics; 2) score normalization and revision theory; 3) face selection and tracking; and 4) identity management, are described here using an integrated approach that includes transduction and boosting for ranking and sequential fusion/aggregation, respectively, on one side, and active learning and change/ outlier/intrusion detection realized using information gain and martingale, respectively, on the other side. The methodology proposed can be mapped to additional types of information beyond biometrics.

Single-wafer die-level fusion bonding process for multi-layer devices

Single-wafer die-level fusion bonding (SDFB) process is proposed, which can reduce the mask number and fabrication time in the prototype research of multi-layer fusion bonding devices. The conception and scheme are illustrated using micro gas turbine of MIT as an example. The techniques to etch different mask and depth in a single-wafer are listed and analyzed. The alignment compensation mechanism in the SDFB is also discussed which can avoid the photolithography misalignment and improve the precision between key layers. The advantages and disadvantages of SDFB are analyzed in detail. The feasibility of the scheme is verified by the experiment. Besides micro gas turbine of MIT, the SDFB process can also be used in other multi-layer fusion bonding devices such as ultra thick proof mass inertial sensors, multi-layer fluidic devices, etc.

Kalman Fuser with Multi-layer Fusion Structure for Descriptor Systems

摘要: 对带多传感器的线性离散随机广义系统, 用奇异值分解将其化为两个降阶耦合子系统, 应用现代时间序列分析方法, 基于自回归滑动平均 (Autoregressive moving average, ARMA) 新息模型和白噪声估计理论, 提出了带三层融合结构的分布式稳态 Kalman 融合器, 它由两个加权融合器和两个复合融合器组成. 第一层给出子系统状态融合器, 实现了每个子系统分量解耦融合; 第二层给出变换后状态融合器, 实现了两个子系统的解耦融合; 第三层给出原始状态融合器, 它可统一处理状态融合滤波、平滑和预报问题. 为计算最优加权阵, 给出了计算局部估计误差互协方差阵公式, 证明了它的精度比每个局部估值器精度高. Monte Carlo 的仿真实例说明了其有效性. 关键词: 多传感器信息融合 / 广义系统 / Kalman滤波器 / 多层融合 / 解耦融合 / 现代时间序列分析方法

Determination of the fracture toughness of inhomogeneous three-layer fusion claddings on compact specimens

We propose a method for the evaluation of the fracture toughness (K1C) of multilayer fusion claddings on compact specimens at elevated temperatures (20, 70, and 300°C) with cladding (in the slot) of three layers of materials with gradually increasing crack resistance. As the testing temperature increases, the value of K1C decreases for harder and stronger fusion claddings and, vise versa, increases for less strong and softer materials.

Information Fusion for Computational Assessment of Air Quality and Health Effects

SM5-PD-07 This work presents the functional layout, the operational features and characteristic results of the information fusion method ICAROS NET. This is an innovative technique for the assessment of air quality and the related potential health effects at the urban and regional scales. It is based on multilayer fusion of environmental and epidemiologic data and models aiming at reducing the error inherent to environmental measurements and their statistical interpretation. ICAROS NET exploits to the fullest the information potential of Earth observation data, atmospheric chemical and transport models, and ground-based measurements. The assimilation of information from all 3 data sources into an optimized computational model allows the estimation of tropospheric particulate loading at very high precision and very high spatial resolution. The computational system and the corresponding method for human exposure and health effect assessment due to airborne toxicants is based on: Integration and assimilation of different environmental data types, including data coming from satellite-based remote sensing, ground-based air quality measurements, and advanced atmospheric modeling; Emergence of meta-knowledge through the use of inference techniques for systemic data validation and forecasting; Sharing of the knowledge acquired through environmental data fusion among urban policy-makers and concerned stakeholders using current and future telematics infrastructure; Decision support for air quality management at the urban and trans-boundary scales based on multicriteria analysis and interactive optimization. ICAROS NET technology shows great potential for integrated air quality and health management. The information fusion methodology implemented herein functionally integrates all 3 information sources (Earth observation, modelling and analytic measurements) and reduces the overall error of the method to circa 10% with regard to estimation of human exposure to airborne toxicants. Results from the implementation of this method for advanced spatially resolved assessment of human exposure and the associated health effects at the urban (eg, Athens, Greece) and regional (eg, Lombardy, Italy) scales are given. This overall system performance is a significant improvement over the best atmospheric models and the pollutant concentration maps produced by spatial interpolation of measurements from the ground. Based on the increased accuracy of the result of information fusion in terms of air quality characterization and taking advantage of its specially designed geo-database, ICAROS NET provides an equally powerful tool for calculating the relative risk on human health posed by airborne chemicals in fine particle form. On the basis of average values of PM loading in Athens, acute mortality due to airborne particles was estimated to have increased by 0.1% to 1% in 2002 and by 1% to 6% (in the city center) in 2003. Chronic mortality values were significantly higher ranging from 10% increase in 2002 up to 30% (in some hot spots) in 2003. Morbidity indicators such as increase in hospital admissions with respiratory problems showed a similarly increasing trend. Analogous results at the regional scale in Lombardy, northern Italy, showed a definitively increasing trend in both morbidity and mortality indicators, albeit more contained than in the case of Athens for the period 2001 to 2003.

A TDOA location algorithm based on data fusion

A new positioning method in mobile networks is presented. Based on the data fusion technology, it processes multi-layer information fusion for the location estimates achieved by the Chan algorithm, which increases mobile positioning accuracy effectively by only using measured difference of arriving (TDOA) signals. The method is simple and practical, especially when the location estimates are corrupted by the non-line-of-sight (NLOS) error. It not only has high positioning accuracy, but also reduces the location failure probability. Results from computer simulation show that the proposed method is effective in various environments.

Information Fusion for Computational Assessment of Air Quality and Health Effects

Satellite-derived data can be used to augment ground-based measurements of air quality and health effects. This article discusses the functional layout, the operational features, and the characteristic results of the information fusion method ICAROS NET. This method is based on the multilayer fusion of environmental and epidemiological data and models. The authors stress that ICAROS NET exploits to the fullest the information potential of Earth observation data, atmospheric chemical and transport models, and ground-based measurements. The primary strength of ICAROS NET is the database. The authors conclude that the assimilation of information from all three data sources into an optimized computational model allows the estimation of tropospheric particulate loading at very high precision and very high spatial resolution. The authors use a case study of historical information from Athens, Greece to illustrate the concepts under discussion.