skip to main content


Title: An Accelerated Rank-(L,L,1,1) Block Term Decomposition Of Multi-Subject Fmri Data Under Spatial Orthonormality Constraint
The decomposition of multi-subject fMRI data using rank- (L,L,1,1) block term decomposition (BTD) can preserve higher-way data structure and is more robust to noise effects by decomposing shared spatial maps (SMs) into a product of two rank-L loading matrices. However, since the number of whole-brain voxels is very large and rank L is larger than 1, the rank-(L,L,1,1) BTD requires high computation and memory. Therefore, we propose an accelerated rank- (L,L,1,1) BTD algorithm based upon the method of alternating least squares (ALS). We speed up updates of loading matrices by reducing fMRI data into subspaces, and add an orthonormality constraint on shared SMs to improve the performance. Moreover, we evaluate the rank-L effect on the proposed method for actual task-related fMRI data. The proposed method shows better performance when L=35. Meanwhile, experimental comparison results verify that the proposed method largely reduced (17.36 times) computation time compared to ALS while also providing satisfying separation performance.  more » « less
Award ID(s):
2112455
NSF-PAR ID:
10331814
Author(s) / Creator(s):
; ; ; ; ; ; ;
Date Published:
Journal Name:
The International Conference on Acoustics, Speech, & Signal Processing (ICASSP)
Page Range / eLocation ID:
3933 to 3937
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Matrices are exceptionally useful in various fields of study as they provide a convenient framework to organize and manipulate data in a structured manner. However, modern matrices can involve billions of elements, making their storage and processing quite demanding in terms of computational resources and memory usage. Although prohibitively large, such matrices are often approximately low rank. We propose an algorithm that exploits this structure to obtain a low rank decomposition of any matrix A as A≈LR, where L and R are the low rank factors. The total number of elements in L and R can be significantly less than that in A. Furthermore, the entries of L and R are quantized to low precision formats −− compressing A by giving us a low rank and low precision factorization. Our algorithm first computes an approximate basis of the range space of A by randomly sketching its columns, followed by a quantization of the vectors constituting this basis. It then computes approximate projections of the columns of A onto this quantized basis. We derive upper bounds on the approximation error of our algorithm, and analyze the impact of target rank and quantization bit-budget. The tradeoff between compression ratio and approximation accuracy allows for flexibility in choosing these parameters based on specific application requirements. We empirically demonstrate the efficacy of our algorithm in image compression, nearest neighbor classification of image and text embeddings, and compressing the layers of LlaMa-7b. Our results illustrate that we can achieve compression ratios as aggressive as one bit per matrix coordinate, all while surpassing or maintaining the performance of traditional compression techniques. 
    more » « less
  2. Purpose

    To develop a physics‐guided deep learning (PG‐DL) reconstruction strategy based on a signal intensity informed multi‐coil (SIIM) encoding operator for highly‐accelerated simultaneous multislice (SMS) myocardial perfusion cardiac MRI (CMR).

    Methods

    First‐pass perfusion CMR acquires highly‐accelerated images with dynamically varying signal intensity/SNR following the administration of a gadolinium‐based contrast agent. Thus, using PG‐DL reconstruction with a conventional multi‐coil encoding operator leads to analogous signal intensity variations across different time‐frames at the network output, creating difficulties in generalization for varying SNR levels. We propose to use a SIIM encoding operator to capture the signal intensity/SNR variations across time‐frames in a reformulated encoding operator. This leads to a more uniform/flat contrast at the output of the PG‐DL network, facilitating generalizability across time‐frames. PG‐DL reconstruction with the proposed SIIM encoding operator is compared to PG‐DL with conventional encoding operator, split slice‐GRAPPA, locally low‐rank (LLR) regularized reconstruction, low‐rank plus sparse (L + S) reconstruction, and regularized ROCK‐SPIRiT.

    Results

    Results on highly accelerated free‐breathing first pass myocardial perfusion CMR at three‐fold SMS and four‐fold in‐plane acceleration show that the proposed method improves upon the reconstruction methods use for comparison. Substantial noise reduction is achieved compared to split slice‐GRAPPA, and aliasing artifacts reduction compared to LLR regularized reconstruction, L + S reconstruction and PG‐DL with conventional encoding. Furthermore, a qualitative reader study indicated that proposed method outperformed all methods.

    Conclusion

    PG‐DL reconstruction with the proposed SIIM encoding operator improves generalization across different time‐frames /SNRs in highly accelerated perfusion CMR.

     
    more » « less
  3. Abstract

    Geostatistical modeling for continuous point‐referenced data has extensively been applied to neuroimaging because it produces efficient and valid statistical inference. However, diffusion tensor imaging (DTI), a neuroimaging technique characterizing the brain's anatomical structure, produces a positive‐definite (p.d.) matrix for each voxel. Currently, only a few geostatistical models for p.d. matrices have been proposed because introducing spatial dependence among p.d. matrices properly is challenging. In this paper, we use the spatial Wishart process, a spatial stochastic process (random field), where each p.d. matrix‐variate random variable marginally follows a Wishart distribution, and spatial dependence between random matrices is induced by latent Gaussian processes. This process is valid on an uncountable collection of spatial locations and is almost‐surely continuous, leading to a reasonable way of modeling spatial dependence. Motivated by a DTI data set of cocaine users, we propose a spatial matrix‐variate regression model based on the spatial Wishart process. A problematic issue is that the spatial Wishart process has no closed‐form density function. Hence, we propose an approximation method to obtain a feasible Cholesky decomposition model, which we show to be asymptotically equivalent to the spatial Wishart process model. A local likelihood approximation method is also applied to achieve fast computation. The simulation studies and real data application demonstrate that the Cholesky decomposition process model produces reliable inference and improved performance, compared to other methods.

     
    more » « less
  4. null (Ed.)
    We consider the problem of low-rank approximation of massive dense nonnegative tensor data, for example, to discover latent patterns in video and imaging applications. As the size of data sets grows, single workstations are hitting bottlenecks in both computation time and available memory. We propose a distributed-memory parallel computing solution to handle massive data sets, loading the input data across the memories of multiple nodes, and performing efficient and scalable parallel algorithms to compute the low-rank approximation. We present a software package called Parallel Low-rank Approximation with Nonnegativity Constraints, which implements our solution and allows for extension in terms of data (dense or sparse, matrices or tensors of any order), algorithm (e.g., from multiplicative updating techniques to alternating direction method of multipliers), and architecture (we exploit GPUs to accelerate the computation in this work). We describe our parallel distributions and algorithms, which are careful to avoid unnecessary communication and computation, show how to extend the software to include new algorithms and/or constraints, and report efficiency and scalability results for both synthetic and real-world data sets. 
    more » « less
  5. Tucker decomposition is a low-rank tensor approximation that generalizes a truncated matrix singular value decomposition (SVD). Existing parallel software has shown that Tucker decomposition is particularly effective at compressing terabyte-sized multidimensional scientific simulation datasets, computing reduced representations that satisfy a specified approximation error. The general approach is to get a low-rank approximation of the input data by performing a sequence of matrix SVDs of tensor unfoldings, which tend to be short-fat matrices. In the existing approach, the SVD is performed by computing the eigendecomposition of the Gram matrix of the unfolding. This method sacrifices some numerical stability in exchange for lower computation costs and easier parallelization. We propose using a more numerically stable though more computationally expensive way to compute the SVD by pre- processing with a QR decomposition step and computing an SVD of only the small triangular factor. The more numerically stable approach allows us to achieve the same accuracy with half the working precision (for example, single rather than double precision). We demonstrate that our method scales as well as the existing approach, and the use of lower precision leads to an overall reduction in running time of up to a factor of 2 when using 10s to 1000s of processors. Using the same working precision, we are also able to compute Tucker decompositions with much smaller approximation error. 
    more » « less