21,907 Hits in 4.8 sec

Identifiable Phenotyping using Constrained Non-Negative Matrix Factorization [article]

Shalmali Joshi, Suriya Gunasekar, David Sontag, Joydeep Ghosh
2016 arXiv   pre-print
A basic latent factor estimation technique of non-negative matrix factorization (NMF) is augmented with domain specific constraints to obtain sparse latent factors that are anchored to a fixed set of chronic  ...  The proposed method can be readily adapted to any non-negative EHR data across various healthcare institutions.  ...  We also thank Yacine Jernite for sharing a code used in preprocessing clinical notes.  ... 
arXiv:1608.00704v3 fatcat:lwhnqjk6wnhbvc2rnsw3wjv75u


Ardavan Afshar, Ioakeim Perros, Haesun Park, Christopher deFilippi, Xiaowei Yan, Walter Stewart, Joyce Ho, Jimeng Sun
2020 Proceedings of the ACM Conference on Health, Inference, and Learning  
TASTE combines the PARAFAC2 model with non-negative matrix factorization to model a temporal and a static tensor.  ...  Using 60 phenotypes extracted by TASTE, a simple logistic regression can achieve the same level of area under the curve (AUC) for HF prediction compared to a deep learning model using recurrent neural  ...  Therefore, we consider as an additional factor matrix, constrain it to be non-negative, and minimize its difference to .  ... 
doi:10.1145/3368555.3384464 pmid:33659966 pmcid:PMC7924914 dblp:conf/chil/AfsharPPdYSHS20 fatcat:mkqiyz2invftxjaqyyrmbftwcu

COPA: Constrained PARAFAC2 for Sparse & Large Datasets [article]

Ardavan Afshar, Ioakeim Perros, Evangelos E. Papalexakis, Elizabeth Searles, Joyce Ho, Jimeng Sun
2018 arXiv   pre-print
To tackle these challenges, we propose a COnstrained PARAFAC2 (COPA) method, which carefully incorporates optimization constraints such as temporal smoothness, sparsity, and non-negativity in the resulting  ...  factors.  ...  Non-negativity on S k : COPA is able to impose non-negativity constraint to factor matrices H, S k , and V .  ... 
arXiv:1803.04572v2 fatcat:d4yu5jqq2rcgxmp6xmuc27pufy

Integrating Hypertension Phenotype and Genotype with Hybrid Non-negative Matrix Factorization [article]

Yuan Luo, Chengsheng Mao, Yiben Yang, Fei Wang, Faraz S. Ahmad, Donna Arnett, Marguerite R. Irvin, Sanjiv J. Shah
2018 arXiv   pre-print
We aim to provide informed patient stratification by introducing Hybrid Non-negative Matrix Factorization (HNMF) on phenotype and genotype matrices.  ...  On real-world clinical dataset, we used the patient factor matrix as features to predict main cardiac mechanistic outcomes.  ...  From the method perspective, Non-negative Matrix Factorization (NMF) refers to the set of problems on approximating a non-negative matrix as the product of several non-negative matrices.  ... 
arXiv:1805.05008v2 fatcat:4vson5l26jc4ridpdlng76ueeu

TASTE: Temporal and Static Tensor Factorization for Phenotyping Electronic Health Records [article]

Ardavan Afshar, Ioakeim Perros, Haesun Park, Christopher deFilippi, Xiaowei Yan, Walter Stewart, Joyce Ho, Jimeng Sun
2019 arXiv   pre-print
TASTE combines the PARAFAC2 model with non-negative matrix factorization to model a temporal and a static tensor.  ...  Using 80 phenotypes extracted by TASTE, a simple logistic regression can achieve the same level of area under the curve (AUC) for HF prediction compared to a deep learning model using recurrent neural  ...  Therefore, we consider U k as an additional factor matrix, constrain it to be non-negative and minimize its difference to Q k H.  ... 
arXiv:1911.05843v1 fatcat:butmvfds35bpnjvz6uxxrq432a

Mining Functional Modules by Multiview-NMF of Phenome-Genome Association [article]

YaoGong Zhang, YingJie Xu, Xin Fan, YuXiang Hong, Jiahui Liu, ZhiCheng He, YaLou Huang, MaoQiang Xie
2017 arXiv   pre-print
Results: We propose a hierarchal Nonnegative Matrix Factorization (NMF)-based method, called Consistent Multiple Nonnegative Matrix Factorization (CMNMF), to factorize genome-phenome association matrix  ...  In this work, we explore the plausibility of detecting gene modules by factorizing gene-phenotype associations from a phenotype ontology rather than the conventionally used gene expression data.  ...  Based on this motivation, we propose a multi-view NMF-based method called CMNMF (consistent multiple non-negative matrix factorization) for mining functional gene modules, in which the hierarchical structure  ... 
arXiv:1705.03998v1 fatcat:6f4lbwffhzfqxl2pvlplknstiy

Phenotyping using Structured Collective Matrix Factorization of Multi--source EHR Data [article]

Suriya Gunasekar, Joyce C. Ho, Joydeep Ghosh, Stephanie Kreml, Abel N Kho, Joshua C Denny, Bradley A Malin, Jimeng Sun
2016 arXiv   pre-print
., diagnosis, medications, and lab reports) available in heterogeneous datatypes in a generalized Collective Matrix Factorization (CMF), our methods can generate rich phenotypes.  ...  We propose a constrained formulation of CMF for estimating sparse phenotypes.  ...  Secondly, in the current framework, we post-process the candidate phenotypes to enforce hard sparsity requirements.  ... 
arXiv:1609.04466v1 fatcat:rbjfhryg3jdk7lljpnvdrzbi3m

Enter the matrix: factorization uncovers knowledge from omics [article]

Genevieve L. Stein-O'Brien, Raman Arora, Aedin C. Culhane, Alexander Favorov, Lana X. Garmire, Casey Greene, Loyal A. Goff, Yifeng Li, Alioune Ngom, Michael F. Ochs, Yanxun Xu, Elana J. Fertig
2017 bioRxiv   pre-print
We review exemplary applications of matrix factorization for systems-level analyses.  ...  Matrix factorization techniques can reveal low-dimensional structure from high-dimensional data that reflect these interactions.  ...  Non-negative matrix factorization (NMF) is a group of algorithms that constrains all elements of the ܷ and ܸ matrices to be greater than or equal to zero.  ... 
doi:10.1101/196915 fatcat:booyynhuazekvhox2ey6x2s4oy

Unsupervised EHR-based Phenotyping via Matrix and Tensor Decompositions [article]

Florian Becker, Age K. Smilde, Evrim Acar
2022 arXiv   pre-print
Low-rank data approximation methods such as matrix (e.g., non-negative matrix factorization) and tensor decompositions (e.g., CANDECOMP/PARAFAC) have demonstrated that they can provide such transparent  ...  The existing literature is categorized into temporal vs. static phenotyping approaches based on matrix vs. tensor decompositions.  ...  By imposing non-negativity constraints on the factor matrices, the problem can be formulated as a non-negative matrix factorization (NMF) (Lee & Seung, 1999) problem.  ... 
arXiv:2209.00322v1 fatcat:xqziwv3ifjeo7njle5xo4zfocy

Identification of Tumor Subtypes of Endometrial Carcinoma by Integration of Heterogeneous Datasets

Kim H, Bredel M
2015 Journal of Medical Diagnostic Methods  
Although several methodological approaches have been proposed for the heterogeneous data integration, there is no framework of sparse non-negative matrix factorization (NMF) for handling heterogeneous  ...  factor target network.  ...  Clustering methods based on matrix computations, such as non-negative matrix factorization (NMF), can be modified to deal with this complex problem.  ... 
doi:10.4172/2168-9784.1000189 fatcat:qkmx5soo7nf55gxu7z7rxrvbpi

Integrating hypertension phenotype and genotype with hybrid non-negative matrix factorization

Yuan Luo, Chengsheng Mao, Yiben Yang, Fei Wang, Faraz S Ahmad, Donna Arnett, Marguerite R Irvin, Sanjiv J Shah, Jonathan Wren
2018 Bioinformatics  
Hypertension is a heterogeneous syndrome in need of improved subtyping using phenotypic and genetic measurements with the goal of identifying subtypes of patients who share similar pathophysiologic mechanisms  ...  We aim to provide informed patient stratification based on phenotype and genotype features.  ...  HNMF stands for hybrid non-negative matrix factorization.LGD stands for likely gene disruptive Fig. 2 . 2 Hybrid non-negative matrix factorization model.  ... 
doi:10.1093/bioinformatics/bty804 pmid:30239588 pmcid:PMC6477985 fatcat:l2vc3nrma5fvvbfcqrsckxxhzi

Inferring Multidimensional Rates of Aging from Cross-Sectional Data [article]

Emma Pierson, Pang Wei Koh, Tatsunori Hashimoto, Daphne Koller, Jure Leskovec, Nicholas Eriksson, Percy Liang
2019 arXiv   pre-print
We prove that when this nonlinear function is constrained to be order-isomorphic, the model family is identifiable solely from cross-sectional data provided the distribution of time-independent variation  ...  On the UK Biobank human health dataset, our model reconstructs the observed data while learning interpretable rates of aging associated with diseases, mortality, and aging risk factors.  ...  Applying Lemma 3 gives us that J q (z) is a non-negative monomial matrix.  ... 
arXiv:1807.04709v3 fatcat:jobkallrnzejnd2j32twp4qlau


Yichen Wang, Robert Chen, Joydeep Ghosh, Joshua C. Denny, Abel Kho, You Chen, Bradley A. Malin, Jimeng Sun
2015 Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD '15  
We propose Rubik, a constrained non-negative tensor factorization and completion method for phenotyping.  ...  In particular, by using knowledge guidance constraints, Rubik can also discover sub-phenotypes for several major diseases.  ...  [18] proposed an alternating non-negative least square method with a block pivoting technique. Constrained Factorization.  ... 
doi:10.1145/2783258.2783395 pmid:31452969 pmcid:PMC6709413 fatcat:2giepnzwk5exze34bjywwnyc4e

Constrained tensor factorization for computational phenotyping and mortality prediction in patients with cancer [article]

Francisco Y Cai, Chengsheng Mao, Yuan Luo
2021 arXiv   pre-print
EHR data, represented as a three-dimensional analogue of a matrix (tensor), is decomposed into two-dimensional factors that can be interpreted as computational phenotypes.  ...  Conclusion: Constrained tensor factorization, applied to sparse EHR data of patients with cancer, can discover computational phenotypes predictive of five-year mortality.  ...  The utility of incorporating SDOH confounding variables can be achieved by extending "confounding-aware" non-negative matrix factorization [23] to tensor factorization [7, 28] .  ... 
arXiv:2112.12933v1 fatcat:luwl3oibpjffpcmpnymny5yxwm

Decomposing Group Differences of Latent Means of Ordered Categorical Variables within a Genetic Factor Model

Seung Bin Cho, Phillip K. Wood, Andrew C. Heath
2008 Behavior Genetics  
A genetic factor model is introduced for decomposition of group differences of the means of phenotypic behavior as well as individual differences when the research variables under consideration are ordered  ...  Use of the proposed model is illustrated using a measure of conservatism in the data collected from the Australian Twin Registry.  ...  Constraint (c) identifies the factor loadings for the non-reference groups.  ... 
doi:10.1007/s10519-008-9237-9 pmid:19009342 pmcid:PMC3401167 fatcat:uli6kpkdnngs3kyj5r6teuorxm
« Previous Showing results 1 — 15 out of 21,907 results