A Hybrid Protection Scheme for the Gait Analysis in Early Dementia Recognition

Human activity recognition (HAR) through gait analysis is a very promising research area for early detection of neurodegenerative diseases because gait abnormalities are typical symptoms of some neurodegenerative diseases, such as early dementia. While working with such biometric data, the performance parameters must be considered along with privacy and security issues. In other words, such biometric data should be processed under specific security and privacy requirements. This work proposes an innovative hybrid protection scheme combining a partially homomorphic encryption scheme and a cancelable biometric technique based on random projection to protect gait features, ensuring patient privacy according to ISO/IEC 24745. The proposed hybrid protection scheme has been implemented along a long short-term memory (LSTM) neural network to realize a secure early dementia diagnosis system. The proposed protection scheme is scalable and implementable with any type of neural network because it is independent of the network's architecture. The conducted experiments demonstrate that the proposed protection scheme enables a high trade-off between safety and performance. The accuracy degradation is at most 1.20% compared with the early dementia recognition system without the protection scheme. Moreover, security and computational analyses of the proposed scheme have been conducted and reported.

PGAGP: Predicting pathogenic genes based on adaptive network embedding algorithm

The study of disease-gene associations is an important topic in the field of computational biology. The accumulation of massive amounts of biomedical data provides new possibilities for exploring potential relations between diseases and genes through computational strategy, but how to extract valuable information from the data to predict pathogenic genes accurately and rapidly is currently a challenging and meaningful task. Therefore, we present a novel computational method called PGAGP for inferring potential pathogenic genes based on an adaptive network embedding algorithm. The PGAGP algorithm is to first extract initial features of nodes from a heterogeneous network of diseases and genes efficiently and effectively by Gaussian random projection and then optimize the features of nodes by an adaptive refining process. These low-dimensional features are used to improve the disease-gene heterogenous network, and we apply network propagation to the improved heterogenous network to predict pathogenic genes more effectively. By a series of experiments, we study the effect of PGAGP's parameters and integrated strategies on predictive performance and confirm that PGAGP is better than the state-of-the-art algorithms. Case studies show that many of the predicted candidate genes for specific diseases have been implied to be related to these diseases by literature verification and enrichment analysis, which further verifies the effectiveness of PGAGP. Overall, this work provides a useful solution for mining disease-gene heterogeneous network to predict pathogenic genes more effectively.

[A discrimination model for differentiation of renal cell carcinoma from renal angiomyolipoma without visible fat: based on hierarchical fusion framework of multi-classifier]

OBJECTIVE

To investigate the capabilities of classification models based on hierarchical fusion framework of multi-classifier using a random projection strategy for differentiation of renal cell carcinoma (RCC) from small renal angiomyolipoma (< 4 cm) without visible fat (AMLwvf).

METHODS

We retrospectively collected the clinical data from 163 patients with pathologically proven small renal mass, including 118 with RCC and 45 with AMLwvf.Target region of interest (ROI) delineation was performed on an unenhanced phase (UP) CT image slice displaying the largest lesion area.The radiomics features were used to establish a hierarchical fusion method.On the projection-based level, the homogeneous classifiers were fused, and the fusion results were further fused at the classifier-based level to construct a multi-classifier fusion system based on random projection for differentiation of AMLwvf and RCC.The discriminative capability of this model was quantitatively evaluated using 5-fold cross validation and 4 evaluation indexes[specificity, sensitivity, accuracy and area under ROC curve (AUC)].We quantitatively compared this multi-classifier fusion framework against different classification models using a single classifier and several multi-classifier ensemble models.

RESULTS

When the projection number was set at 10, the proposed hierarchical fusion differentiation framework achieved the best results on all the evaluation measurements.At the optimal projection number of 10, the specificity, sensitivity, average accuracy and AUC of the multi-classifier ensemble classification system for differentiation between AMLwvf and RCC were 0.853, 0.693, 0.809 and 0.870, respectively.

CONCLUSION

The proposed model constructed based on a multi-classifier fusion system using random projection shows better performance to differentiate RCC from AMLwvf than the AMLwvf and RCC discrimination models based on a single classification algorithm and the currently available benchmark ensemble methods.

How to reduce dimension with PCA and random projections?

In our "big data" age, the size and complexity of data is steadily increasing. Methods for dimension reduction are ever more popular and useful. Two distinct types of dimension reduction are "data-oblivious" methods such as random projections and sketching, and "data-aware" methods such as principal component analysis (PCA). Both have their strengths, such as speed for random projections, and data-adaptivity for PCA. In this work, we study how to combine them to get the best of both. We study "sketch and solve" methods that take a random projection (or sketch) first, and compute PCA after. We compute the performance of several popular sketching methods (random iid projections, random sampling, subsampled Hadamard transform, CountSketch, etc) in a general "signal-plus-noise" (or spiked) data model. Compared to well-known works, our results (1) give asymptotically exact results, and (2) apply when the signal components are only slightly above the noise, but the projection dimension is non-negligible. We also study stronger signals allowing more general covariance structures. We find that (a) signal strength decreases under projection in a delicate way depending on the structure of the data and the sketching method, (b) orthogonal projections are slightly more accurate, (c) randomization does not hurt too much, due to concentration of measure, (d) CountSketch can be somewhat improved by a normalization method. Our results have implications for statistical learning and data analysis. We also illustrate that the results are highly accurate in simulations and in analyzing empirical data.

Sparse and Random Sampling Techniques for High-Resolution, Full-Field, BSS-Based Structural Dynamics Identification from Video

Video-based techniques for identification of structural dynamics have the advantage that they are very inexpensive to deploy compared to conventional accelerometer or strain gauge techniques. When structural dynamics from video is accomplished using full-field, high-resolution analysis techniques utilizing algorithms on the pixel time series such as principal components analysis and solutions to blind source separation the added benefit of high-resolution, full-field modal identification is achieved. An important property of video of vibrating structures is that it is particularly sparse. Typically video of vibrating structures has a dimensionality consisting of many thousands or even millions of pixels and hundreds to thousands of frames. However the motion of the vibrating structure can be described using only a few mode shapes and their associated time series. As a result, emerging techniques for sparse and random sampling such as compressive sensing should be applicable to performing modal identification on video. This work presents how full-field, high-resolution, structural dynamics identification frameworks can be coupled with compressive sampling. The techniques described in this work are demonstrated to be able to recover mode shapes from experimental video of vibrating structures when 70% to 90% of the frames from a video captured in the conventional manner are removed.

A New Look at the Structures of Old Sepsis Actors by Exploratory Data Analysis Tools

Sepsis is a life-threatening condition that accounts for numerous deaths worldwide, usually complications of common community infections (i.e., pneumonia, etc), or infections acquired during the hospital stay. Sepsis and septic shock, its most severe evolution, involve the whole organism, recruiting and producing a lot of molecules, mostly proteins. Proteins are dynamic entities, and a large number of techniques and studies have been devoted to elucidating the relationship between the conformations adopted by proteins and what is their function. Although molecular dynamics has a key role in understanding these relationships, the number of protein structures available in the databases is so high that it is currently possible to build data sets obtained from experimentally determined structures. Techniques for dimensionality reduction and clustering can be applied in exploratory data analysis in order to obtain information on the function of these molecules, and this may be very useful in immunology to better understand the structure-activity relationship of the numerous proteins involved in host defense, moreover in septic patients. The large number of degrees of freedom that characterize the biomolecules requires special techniques which are able to analyze this kind of data sets (with a small number of entries respect to the number of degrees of freedom). In this work we analyzed the ability of two different types of algorithms to provide information on the structures present in three data sets built using the experimental structures of allosteric proteins involved in sepsis. The results obtained by means of a principal component analysis algorithm and those obtained by a random projection algorithm are largely comparable, proving the effectiveness of random projection methods in structural bioinformatics. The usefulness of random projection in exploratory data analysis is discussed, including validation of the obtained clusters. We have chosen these proteins because of their involvement in sepsis and septic shock, aimed to highlight the potentiality of bioinformatics to point out new diagnostic and prognostic tools for the patients.

An Efficient and Robust Unsupervised Anomaly Detection Method Using Ensemble Random Projection in Surveillance Videos

Video anomaly detection is widely applied in modern society, which is achieved by sensors such as surveillance cameras. This paper learns anomalies by exploiting videos under the fully unsupervised setting. To avoid massive computation caused by back-prorogation in existing methods, we propose a novel efficient three-stage unsupervised anomaly detection method. In the first stage, we adopt random projection instead of autoencoder or its variants in previous works. Then we formulate the optimization goal as a least-square regression problem which has a closed-form solution, leading to less computational cost. The discriminative reconstruction losses of normal and abnormal events encourage us to roughly estimate normality that can be further sifted in the second stage with one-class support vector machine. In the third stage, to eliminate the instability caused by random parameter initializations, ensemble technology is performed to combine multiple anomaly detectors' scores. To the best of our knowledge, it is the first time that unsupervised ensemble technology is introduced to video anomaly detection research. As demonstrated by the experimental results on several video anomaly detection benchmark datasets, our algorithm robustly surpasses the recent unsupervised methods and performs even better than some supervised approaches. In addition, we achieve comparable performance contrast with the state-of-the-art unsupervised method with much less running time, indicating the effectiveness, efficiency, and robustness of our proposed approach.

Prediction of Self-Interacting Proteins from Protein Sequence Information Based on Random Projection Model and Fast Fourier Transform

It is significant for biological cells to predict self-interacting proteins (SIPs) in the field of bioinformatics. SIPs mean that two or more identical proteins can interact with each other by one gene expression. This plays a major role in the evolution of protein‒protein interactions (PPIs) and cellular functions. Owing to the limitation of the experimental identification of self-interacting proteins, it is more and more significant to develop a useful biological tool for the prediction of SIPs from protein sequence information. Therefore, we propose a novel prediction model called RP-FFT that merges the Random Projection (RP) model and Fast Fourier Transform (FFT) for detecting SIPs. First, each protein sequence was transformed into a Position Specific Scoring Matrix (PSSM) using the Position Specific Iterated BLAST (PSI-BLAST). Second, the features of protein sequences were extracted by the FFT method on PSSM. Lastly, we evaluated the performance of RP-FFT and compared the RP classifier with the state-of-the-art support vector machine (SVM) classifier and other existing methods on the human and yeast datasets; after the five-fold cross-validation, the RP-FFT model can obtain high average accuracies of 96.28% and 91.87% on the human and yeast datasets, respectively. The experimental results demonstrated that our RP-FFT prediction model is reasonable and robust.

KerNL: Kernel-Based Nonlinear Approach to Parallel MRI Reconstruction

The conventional calibration-based parallel imaging method assumes a linear relationship between the acquired multi-channel k-space data and the unacquired missing data, where the linear coefficients are estimated using some auto-calibration data. In this paper, we first analyze the model errors in the conventional calibration-based methods and demonstrate the nonlinear relationship. Then, a much more general nonlinear framework is proposed for auto-calibrated parallel imaging. In this framework, kernel tricks are employed to represent the general nonlinear relationship between acquired and unacquired k-space data without increasing the computational complexity. Identification of the nonlinear relationship is still performed by solving linear equations. Experimental results demonstrate that the proposed method can achieve reconstruction quality superior to GRAPPA and NL-GRAPPA at high net reduction factors.

Prediction of Protein Hotspots from Whole Protein Sequences by a Random Projection Ensemble System

Hotspot residues are important in the determination of protein-protein interactions, and they always perform specific functions in biological processes. The determination of hotspot residues is by the commonly-used method of alanine scanning mutagenesis experiments, which is always costly and time consuming. To address this issue, computational methods have been developed. Most of them are structure based, i.e., using the information of solved protein structures. However, the number of solved protein structures is extremely less than that of sequences. Moreover, almost all of the predictors identified hotspots from the interfaces of protein complexes, seldom from the whole protein sequences. Therefore, determining hotspots from whole protein sequences by sequence information alone is urgent. To address the issue of hotspot predictions from the whole sequences of proteins, we proposed an ensemble system with random projections using statistical physicochemical properties of amino acids. First, an encoding scheme involving sequence profiles of residues and physicochemical properties from the AAindex1 dataset is developed. Then, the random projection technique was adopted to project the encoding instances into a reduced space. Then, several better random projections were obtained by training an IBk classifier based on the training dataset, which were thus applied to the test dataset. The ensemble of random projection classifiers is therefore obtained. Experimental results showed that although the performance of our method is not good enough for real applications of hotspots, it is very promising in the determination of hotspot residues from whole sequences.

The use of random projections for the analysis of mass spectrometry imaging data

The 'curse of dimensionality' imposes fundamental limits on the analysis of the large, information rich datasets that are produced by mass spectrometry imaging. Additionally, such datasets are often too large to be analyzed as a whole and so dimensionality reduction is required before further analysis can be performed. We investigate the use of simple random projections for the dimensionality reduction of mass spectrometry imaging data and examine how they enable efficient and fast segmentation using k-means clustering. The method is computationally efficient and can be implemented such that only one spectrum is needed in memory at any time. We use this technique to reveal histologically significant regions within MALDI images of diseased human liver. Segmentation results achieved following a reduction in the dimensionality of the data by more than 99% (without peak picking) showed that histologic changes due to disease can be automatically visualized from molecular images.

Fast GRAPPA reconstruction with random projection

PURPOSE

To address the issue of computational complexity in generalized autocalibrating partially parallel acquisition (GRAPPA) when several calibration data are used.

METHOD

GRAPPA requires fully sampled data for accurate calibration with increasing data needed for higher reduction factors to maintain accuracy, which leads to longer computational time, especially in a three-dimensional (3D) setting and with higher channel count coils. Channel reduction methods have been developed to address this issue when massive array coils are used. In this study, the complexity problem was addressed from a different prospective. Instead of compressing to fewer channels, we propose the use of random projections to reduce the dimension of the linear equation in the calibration phase. The equivalence before and after the reduction is supported by the Johnson-Lindenstrauss lemma. The proposed random projection method can be integrated with channel reduction sequentially for even higher computational efficiency.

RESULTS

Experimental results show that GRAPPA with random projection can achieve comparable image quality with much less computational time when compared with conventional GRAPPA without random projection.

CONCLUSION

The proposed random projection method is able to reduce the computational time of GRAPPA, especially in a 3D setting, without compromising the image quality, or to improve the reconstruction quality by allowing more data for calibration when the computational time is a limiting factor. Magn Reson Med 74:71-80, 2015. © 2014 Wiley Periodicals, Inc.

A new hierarchical method for inter-patient heartbeat classification using random projections and RR intervals

BACKGROUND

The inter-patient classification schema and the Association for the Advancement of Medical Instrumentation (AAMI) standards are important to the construction and evaluation of automated heartbeat classification systems. The majority of previously proposed methods that take the above two aspects into consideration use the same features and classification method to classify different classes of heartbeats. The performance of the classification system is often unsatisfactory with respect to the ventricular ectopic beat (VEB) and supraventricular ectopic beat (SVEB).

METHODS

Based on the different characteristics of VEB and SVEB, a novel hierarchical heartbeat classification system was constructed. This was done in order to improve the classification performance of these two classes of heartbeats by using different features and classification methods. First, random projection and support vector machine (SVM) ensemble were used to detect VEB. Then, the ratio of the RR interval was compared to a predetermined threshold to detect SVEB. The optimal parameters for the classification models were selected on the training set and used in the independent testing set to assess the final performance of the classification system. Meanwhile, the effect of different lead configurations on the classification results was evaluated.

RESULTS

Results showed that the performance of this classification system was notably superior to that of other methods. The VEB detection sensitivity was 93.9% with a positive predictive value of 90.9%, and the SVEB detection sensitivity was 91.1% with a positive predictive value of 42.2%. In addition, this classification process was relatively fast.

CONCLUSIONS

A hierarchical heartbeat classification system was proposed based on the inter-patient data division to detect VEB and SVEB. It demonstrated better classification performance than existing methods. It can be regarded as a promising system for detecting VEB and SVEB of unknown patients in clinical practice.

Challenges of Big Data Analysis

Big Data bring new opportunities to modern society and challenges to data scientists. On one hand, Big Data hold great promises for discovering subtle population patterns and heterogeneities that are not possible with small-scale data. On the other hand, the massive sample size and high dimensionality of Big Data introduce unique computational and statistical challenges, including scalability and storage bottleneck, noise accumulation, spurious correlation, incidental endogeneity, and measurement errors. These challenges are distinguished and require new computational and statistical paradigm. This article gives overviews on the salient features of Big Data and how these features impact on paradigm change on statistical and computational methods as well as computing architectures. We also provide various new perspectives on the Big Data analysis and computation. In particular, we emphasize on the viability of the sparsest solution in high-confidence set and point out that exogeneous assumptions in most statistical methods for Big Data can not be validated due to incidental endogeneity. They can lead to wrong statistical inferences and consequently wrong scientific conclusions.

LSRN: A PARALLEL ITERATIVE SOLVER FOR STRONGLY OVER- OR UNDERDETERMINED SYSTEMS

We describe a parallel iterative least squares solver named LSRN that is based on random normal projection. LSRN computes the min-length solution to min _{x∈ℝ ⁿ} ‖Ax - b‖₂, where A ∈ ℝ ^{m × n} with m ≫ n or m ≪ n, and where A may be rank-deficient. Tikhonov regularization may also be included. Since A is involved only in matrix-matrix and matrix-vector multiplications, it can be a dense or sparse matrix or a linear operator, and LSRN automatically speeds up when A is sparse or a fast linear operator. The preconditioning phase consists of a random normal projection, which is embarrassingly parallel, and a singular value decomposition of size ⌈γ min(m, n)⌉ × min(m, n), where γ is moderately larger than 1, e.g., γ = 2. We prove that the preconditioned system is well-conditioned, with a strong concentration result on the extreme singular values, and hence that the number of iterations is fully predictable when we apply LSQR or the Chebyshev semi-iterative method. As we demonstrate, the Chebyshev method is particularly efficient for solving large problems on clusters with high communication cost. Numerical results show that on a shared-memory machine, LSRN is very competitive with LAPACK's DGELSD and a fast randomized least squares solver called Blendenpik on large dense problems, and it outperforms the least squares solver from SuiteSparseQR on sparse problems without sparsity patterns that can be exploited to reduce fill-in. Further experiments show that LSRN scales well on an Amazon Elastic Compute Cloud cluster.

A novel algorithm for finding interspersed repeat regions

The analysis of repeats in the DNA sequences is an important subject in bioinformatics. In this paper, we propose a novel projection-assemble algorithm to find unknown interspersed repeats in DNA sequences. The algorithm employs random projection algorithm to obtain a candidate fragment set, and exhaustive search algorithm to search each pair of fragments from the candidate fragment set to find potential linkage, and then assemble them together. The complexity of our projection-assemble algorithm is nearly linear to the length of the genome sequence, and its memory usage is limited by the hardware. We tested our algorithm with both simulated data and real biology data, and the results show that our projection-assemble algorithm is efficient. By means of this algorithm, we found an un-labeled repeat region that occurs five times in Escherichia coli genome, with its length more than 5,000 bp, and a mismatch probability less than 4%.