umu.sePublications
Change search
Link to record
Permanent link

Direct link
BETA
Yu, Jun
Alternative names
Publications (10 of 86) Show all publications
Zhou, Z. & Yu, J. (2019). Sparse recovery based on q-ratio constrained minimal singular values. Signal Processing, 155, 247-258
Open this publication in new window or tab >>Sparse recovery based on q-ratio constrained minimal singular values
2019 (English)In: Signal Processing, ISSN 0165-1684, E-ISSN 1872-7557, Vol. 155, p. 247-258Article in journal (Refereed) Published
Abstract [en]

We study verifiable sufficient conditions and computable performance bounds for sparse recovery algorithms such as the Basis Pursuit, the Dantzig selector and the Lasso estimator, in terms of a newly defined family of quality measures for the measurement matrices. With high probability, the developed measures for subgaussian random matrices are bounded away from zero as long as the number of measurements is reasonably large. Comparing to the restricted isotropic constant based performance analysis, the arguments in this paper are much more concise and the obtained bounds are tighter. Numerical experiments are presented to illustrate our theoretical results.

Place, publisher, year, edition, pages
Elsevier, 2019
Keywords
Compressive sensing, q-ratio sparsity, q-ratio constrained minimal singular values, Convex–concave procedure
National Category
Signal Processing Probability Theory and Statistics
Research subject
Mathematical Statistics; Signal Processing
Identifiers
urn:nbn:se:umu:diva-152530 (URN)10.1016/j.sigpro.2018.10.002 (DOI)2-s2.0-85054424540 (Scopus ID)
Projects
Statistical modelling and intelligent data sampling in MRI and PET measurements for cancer therapy assessment
Funder
Swedish Research Council, 340-2013-534
Available from: 2018-10-10 Created: 2018-10-10 Last updated: 2018-12-11Bibliographically approved
Bayisa, F., Zhou, Z., Cronie, O. & Yu, J. (2018). Adaptive algorithm for sparse signal recovery.
Open this publication in new window or tab >>Adaptive algorithm for sparse signal recovery
2018 (English)Manuscript (preprint) (Other academic)
Abstract [en]

Spike and slab priors play a key role in inducing sparsity for sparse signal recovery. The use of such priorsresults in hard non-convex and mixed integer programming problems. Most of the existing algorithms to solve the optimization problems involve either simplifying assumptions, relaxations or high computational expenses. We propose a new adaptive alternating direction method of multipliers (AADMM) algorithm to directly solve the presented optimization problem. The algorithm is based on the one-to-onemapping property of the support and non-zero element of the signal. At each step of the algorithm, we update the support by either adding an index to it or removing an index from it and use the alternatingdirection method of multipliers to recover the signal corresponding to the updated support. Experiments on synthetic data and real-world images show that the proposed AADMM algorithm provides superior performance and is computationally cheaper, compared to the recently developed iterative convex refinement (ICR) algorithm.

Publisher
p. 16
Keywords
sparsity, adaptive algorithm, sparse signal recovery, spike and slab priors
National Category
Probability Theory and Statistics Computational Mathematics
Research subject
Mathematical Statistics
Identifiers
urn:nbn:se:umu:diva-146386 (URN)
Funder
Swedish Research Council, 340-2013-534
Available from: 2018-04-07 Created: 2018-04-07 Last updated: 2018-06-09
Jiao, X., Zhang, H., Zheng, J., Yin, Y., Wang, G., Chen, Y., . . . Yu, J. (2018). Comparative analysis of nonlinear growth curve models for Arabidopsis thaliana rosette leaves. Acta Physiologiae Plantarum, 40(6), Article ID 114.
Open this publication in new window or tab >>Comparative analysis of nonlinear growth curve models for Arabidopsis thaliana rosette leaves
Show others...
2018 (English)In: Acta Physiologiae Plantarum, ISSN 0137-5881, E-ISSN 1861-1664, Vol. 40, no 6, article id 114Article in journal (Refereed) Published
Abstract [en]

As a model organism, modeling and analysis of the phenotype of Arabidopsis thaliana (A. thaliana) leaves for a given genotype can help us better understand leaf growth regulation. A. thaliana leaves growth trajectories are to be nonlinear and the leaves contribute most to the above-ground biomass. Therefore, analysis of their change regulation and development of nonlinear growth models can better understand the phenotypic characteristics of leaves (e.g., leaf size) at different growth stages. In this study, every individual leaf size of A. thaliana rosette leaves was measured during their whole life cycle using non-destructive imaging measurement. And three growth models (Gompertz model, logistic model and Von Bertalanffy model) were analyzed to quantify the rosette leaves growth process of A. thaliana. Both graphical (plots of standardized residuals) and numerical measures (AIC, R2 and RMSE) were used to evaluate the fitted models. The results showed that the logistic model fitted better in describing the growth of A. thaliana leaves compared to Gompertz model and Von Bertalanffy model, as it gave higher R2 and lower AIC and RMSE for the leaves of A. thaliana at different growth stages (i.e., early leaf, mid-term leaf and late leaf).

Place, publisher, year, edition, pages
Springer, 2018
Keywords
A. thaliana, Growth model, Leaf area, Akaike’s information criterion, Non-destructive imaging measurement
National Category
Plant Biotechnology Probability Theory and Statistics
Research subject
biomechanics
Identifiers
urn:nbn:se:umu:diva-147921 (URN)10.1007/s11738-018-2686-8 (DOI)000451233600001 ()26811110 (PubMedID)2-s2.0-85047359291 (Scopus ID)
Available from: 2018-05-22 Created: 2018-05-22 Last updated: 2018-12-12Bibliographically approved
Kuljus, K., Bayisa, F., Bolin, D., Lember, J. & Yu, J. (2018). Comparison of hidden Markov chain models and hidden Markov random field models in estimation of computed tomography images. Communications in Statistics: Case Studies, Data Analysis and Applications, 4(1), 46-55
Open this publication in new window or tab >>Comparison of hidden Markov chain models and hidden Markov random field models in estimation of computed tomography images
Show others...
2018 (English)In: Communications in Statistics: Case Studies, Data Analysis and Applications, ISSN 2373-7484, Vol. 4, no 1, p. 46-55Article in journal (Refereed) Published
Abstract [en]

Two principal areas of application for estimated computed tomography (CT) images are dose calculations in magnetic resonance imaging (MRI) based radiotherapy treatment planning and attenuation correction for positron emission tomography (PET)/MRI. The main purpose of this work is to investigate the performance of hidden Markov (chain) models (HMMs) in comparison to hidden Markov random field (HMRF) models when predicting CT images of head. Obtained results suggest that HMMs deserve a further study for investigating their potential in modeling applications, where the most natural theoretical choice would be the class of HMRF models.

Place, publisher, year, edition, pages
Taylor & Francis Group, 2018
Keywords
Computed tomography; hidden Markov model; hidden Markov random field; magnetic resonance imaging; pseudo-CT; radiotherapy
National Category
Probability Theory and Statistics Medical Image Processing
Research subject
Mathematical Statistics
Identifiers
urn:nbn:se:umu:diva-148242 (URN)10.1080/23737484.2018.1473059 (DOI)
Projects
Statistical modelling and intelligent data sampling in MRI and PET measurements for cancer therapy assessment
Funder
Swedish Research Council, 340-2013-5342
Available from: 2018-05-31 Created: 2018-05-31 Last updated: 2018-10-30Bibliographically approved
Wieloch, T., Ehlers, I., Yu, J., Frank, D., Grabner, M., Gessler, A. & Schleucher, J. (2018). Intramolecular 13C analysis of tree rings provides multiple plant ecophysiology signals covering decades. Scientific Reports, 8, Article ID 5048.
Open this publication in new window or tab >>Intramolecular 13C analysis of tree rings provides multiple plant ecophysiology signals covering decades
Show others...
2018 (English)In: Scientific Reports, ISSN 2045-2322, E-ISSN 2045-2322, Vol. 8, article id 5048Article in journal (Refereed) Published
Abstract [en]

Measurements of carbon isotope contents of plant organic matter provide important information in diverse fields such as plant breeding, ecophysiology, biogeochemistry and paleoclimatology. They are currently based on 13C/12C ratios of specific, whole metabolites, but we show here that intramolecular ratios provide higher resolution information. In the glucose units of tree-ring cellulose of 12 tree species, we detected large differences in 13C/12C ratios (>10‰) among carbon atoms, which provide isotopically distinct inputs to major global C pools, including wood and soil organic matter. Thus, considering position-specific differences can improve characterisation of soil-to-atmosphere carbon fluxes and soil metabolism. In a Pinus nigra tree-ring archive formed from 1961 to 1995, we found novel 13C signals, and show that intramolecular analysis enables more comprehensive and precise signal extraction from tree rings, and thus higher resolution reconstruction of plants’ responses to climate change. Moreover, we propose an ecophysiological mechanism for the introduction of a 13C signal, which links an environmental shift to the triggered metabolic shift and its intramolecular 13C signature. In conclusion, intramolecular 13C analyses can provide valuable new information about long-term metabolic dynamics for numerous applications.

National Category
Other Earth and Related Environmental Sciences
Identifiers
urn:nbn:se:umu:diva-145978 (URN)10.1038/s41598-018-23422-2 (DOI)000428033900002 ()29567963 (PubMedID)
Funder
Knut and Alice Wallenberg Foundation, 2015.0047
Available from: 2018-03-24 Created: 2018-03-24 Last updated: 2018-06-09Bibliographically approved
Zhou, Z. & Yu, J. (2018). On q-ratio CMSV for sparse recovery.
Open this publication in new window or tab >>On q-ratio CMSV for sparse recovery
2018 (English)Manuscript (preprint) (Other academic)
Abstract [en]

Sparse recovery aims to reconstruct an unknown spare or approximately sparse signal from significantly few noisy incoherent linear measurements. As a kind of computable incoherence measure of the measurement matrix, q-ratio constrained minimal singular values (CMSV) was proposed in Zhou and Yu (2018) to derive the performance bounds for sparse recovery. In this paper, we study the geometrical property of the q-ratio CMSV, based on which we establish new sufficient conditions for signal recovery involving both sparsity defect and measurement error. The ℓ1-truncated set q-width of the measurement matrix is developed as the geometrical characterization of q-ratio CMSV.

In addition, we show that the q-ratio CMSVs of a class of structured random matrices are bounded away from zero with high probability as long as the number of measurements is large enough, therefore satisfy those established sufficient conditions. Overall, our results generalize the results in Zhang and Cheng (2012) from q=2 to any q∈(1,∞] and complement the arguments of q-ratio CMSV from a geometrical view.

Publisher
p. 5
Keywords
Sparse recovery; q-ratio sparsity; q-ratio constrained minimal singular values; ℓ1-truncated set q-width.
National Category
Computational Mathematics Probability Theory and Statistics
Research subject
Mathematical Statistics
Identifiers
urn:nbn:se:umu:diva-148241 (URN)
Projects
Statistical modelling and intelligent data sampling in MRI and PET measurements for cancer therapy assessment
Funder
Swedish Research Council, 340-2013-534
Available from: 2018-05-31 Created: 2018-05-31 Last updated: 2018-06-09
Zhou, Z. & Yu, J. (2018). Sparse recovery based on q-ratio constrained minimal singular values.
Open this publication in new window or tab >>Sparse recovery based on q-ratio constrained minimal singular values
2018 (English)Manuscript (preprint) (Other academic)
Abstract [en]

We study verifiable sufficient conditions and computable performance bounds for sparse recovery algorithms such as the Basis Pursuit, the Dantzig selector and the Lasso estimator, in terms of a newly defined family of quality measures for the measurement matrices. With high probability, the developed measures for subgaussian random matrices are bounded away from zero as long as the number of measurements is reasonably large. Comparing to the restricted isotropic constant based performance analysis, the arguments in this paper are much more concise and the obtained bounds are tighter. Numerical experiments are presented to illustrate our theoretical results.

Publisher
p. 26
Keywords
Compressive sensing; q-ratio sparsity; q-ratio constrained minimal singular values; Convex-concave procedure.
National Category
Probability Theory and Statistics Computational Mathematics
Research subject
Mathematical Statistics
Identifiers
urn:nbn:se:umu:diva-144131 (URN)
Projects
Statistical modelling and intelligent data sampling in MRI and PET measurements for cancer therapy assessment
Funder
Swedish Research Council, 340-2013-534
Available from: 2018-01-22 Created: 2018-01-22 Last updated: 2018-06-09
Bayisa, F., Liu, X., Garpebring, A. & Yu, J. (2018). Statistical learning in computed tomography image estimation. Medical physics (Lancaster), 45(12), 5450-5460
Open this publication in new window or tab >>Statistical learning in computed tomography image estimation
2018 (English)In: Medical physics (Lancaster), ISSN 0094-2405, Vol. 45, no 12, p. 5450-5460Article in journal (Refereed) Published
Abstract [en]

Purpose: There is increasing interest in computed tomography (CT) image estimations from magneticresonance (MR) images. The estimated CT images can be utilized for attenuation correction, patientpositioning, and dose planning in diagnostic and radiotherapy workflows. This study aims to introducea novel statistical learning approach for improving CT estimation from MR images and to compare theperformance of our method with the existing model-based CT image estimation methods.

Methods: The statistical learning approach proposed here consists of two stages. At the trainingstage, prior knowledge about tissue types from CT images was used together with a Gaussian mixturemodel (GMM) to explore CT image estimations from MR images. Since the prior knowledge is notavailable at the prediction stage, a classifier based on RUSBoost algorithm was trained to estimatethe tissue types from MR images. For a new patient, the trained classifier and GMMs were used topredict CT image from MR images. The classifier and GMMs were validated by using voxel-leveltenfold cross-validation and patient-level leave-one-out cross-validation, respectively.

Results: The proposed approach has outperformance in CT estimation quality in comparison withthe existing model-based methods, especially on bone tissues. Our method improved CT image estimationby 5% and 23% on the whole brain and bone tissues, respectively.

Conclusions: Evaluation of our method shows that it is a promising method to generate CTimage substitutes for the implementation of fully MR-based radiotherapy and PET/MRI applications

Place, publisher, year, edition, pages
John Wiley & Sons, 2018
Keywords
Computed tomography, CT image estimation, Gaussian mixture model, magnetic resonance imaging, supervised learning
National Category
Probability Theory and Statistics Medical Image Processing
Research subject
Mathematical Statistics
Identifiers
urn:nbn:se:umu:diva-153283 (URN)DOI:10.1002/mp.13204 (DOI)
Projects
Statistical modelling and intelligent data sampling in MRI and PET measurements for cancer therapy assessment
Funder
Swedish Research Council, 340-2013-5342
Available from: 2018-11-15 Created: 2018-11-15 Last updated: 2018-12-14Bibliographically approved
Wieloch, T., Ehlers, I., Yu, J., Frank, D., Grabner, M., Gessler, A. & Schleucher, J. (2018). Tree-ring cellulose exhibits several interannual 13C signals on the intramolecular level. In: Geophysical Research Abstracts: . Paper presented at EGU General Assembly 2018. , 20, Article ID EGU2018-17509-2.
Open this publication in new window or tab >>Tree-ring cellulose exhibits several interannual 13C signals on the intramolecular level
Show others...
2018 (English)In: Geophysical Research Abstracts, 2018, Vol. 20, article id EGU2018-17509-2Conference paper, Poster (with or without abstract) (Refereed)
Abstract
National Category
Natural Sciences
Identifiers
urn:nbn:se:umu:diva-145982 (URN)
Conference
EGU General Assembly 2018
Available from: 2018-03-24 Created: 2018-03-24 Last updated: 2018-06-09
Söderström, T., Fahlén, J., Ferry, M. & Yu, J. (2017). Athletic ability in childhood and adolescence as a predictor of participation in non-elite sports in young adulthood. Sport in Society: Cultures, Media, Politics, Commerce, 1-18
Open this publication in new window or tab >>Athletic ability in childhood and adolescence as a predictor of participation in non-elite sports in young adulthood
2017 (English)In: Sport in Society: Cultures, Media, Politics, Commerce, ISSN 1743-0437, E-ISSN 1743-0445, p. 1-18Article in journal (Refereed) Epub ahead of print
Abstract [en]

In this article, we contribute to the discussion on factors affecting adult participation in organized sport. To this end, we examine whether explanations regarding sport expertise can also add to the understanding of non-elite-level sport participation in young adulthood. Results from questionnaires (n = 572) revealed that date of birth and early sport debut positively correlated to strong sport performance during childhood, which, in turn, were correlated to strong sport performance and being selected for talent groups during adolescence. Finally, strong sport performance during adolescence was positively correlated to sports club membership as young adults. As relative age effects seem to remain throughout childhood and adolescence, we conclude that the underlying variable that affects the selection process and sport participation in young adulthood is date of birth. The results indicate that being active in sport as young adults is contingent on sport-specific variables previously not investigated in research on sport participation.

Place, publisher, year, edition, pages
Routledge, 2017
National Category
Pedagogy
Identifiers
urn:nbn:se:umu:diva-142708 (URN)10.1080/17430437.2017.1409726 (DOI)
Available from: 2017-12-08 Created: 2017-12-08 Last updated: 2018-06-09Bibliographically approved
Organisations

Search in DiVA

Show all publications