A Multichannel Intraluminal Impedance Gastroesophageal Reflux Characterization Algorithm Based On Sparse Representation

Rasouli, Azra; Rabbani, Hossein; Kermani, Saeed; Raisi, Mostafa; Soheilipour, Maryam; Adibi, Peyman

doi:10.1109/jbhi.2021.3076212

Cited by 1 publication

(4 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The fundamental idea behind SDL-1 methods is that data of a particular class tend to have similar sparse representations. Consequently, the reconstruction error for a test sample will be minimal when the learned sub-dictionary of the appropriate class is applied [38], [40], [41]. Yang et al (2010) proposed a meta-face learning method to learn an adaptive class-specific sub-dictionary for every class, during the training phase [42].…”

Section: Related Workmentioning

confidence: 99%

“…Wei et al (2013) employed a locality regularization term in sparse representation-based classification utilizing 2D-DL to consider data locality [39]. A. R. Kenari et al (2021) presented a 2D-DL-based sparse classification method based on two different definitions of the representation error for the classification of multichannel intraluminal impedance-pH signals [38]. In the first definition, the representation error is obtained based on the minimum mean square error, and in the second one, it is calculated based on the maximum l_ρ-norm (0<ρ≤1).…”

Section: Related Workmentioning

confidence: 99%

“…Subsequently, the learning process is realized by solving a constrained optimization problem through the utilization of the Alternating Direction Method of Multipliers (ADMM) approach [37]. An individual sub-dictionary is learned for each class using the observed data from the corresponding class, to minimize reconstruction errors [38]. During the testing phase, every test sample is reconstructed using all three sub-dictionaries separately, and it is assigned to the class associated with the minimum normalized mean square errors (NMSE) [38].…”

Section: Introductionmentioning

confidence: 99%

“…An individual sub-dictionary is learned for each class using the observed data from the corresponding class, to minimize reconstruction errors [38]. During the testing phase, every test sample is reconstructed using all three sub-dictionaries separately, and it is assigned to the class associated with the minimum normalized mean square errors (NMSE) [38].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

CircWaveDL: Modeling of Optical Coherence Tomography Images Based on a new Supervised Tensor-based Dictionary Learning for Classification of Macular abnormalities

Arian,

Vard,

Kafieh

et al. 2023

Preprint

View full text Add to dashboard Cite

Modeling optical coherence tomography (OCT) images is highly beneficial for various image processing applications as well as assisting ophthalmologists in the early detection of macular abnormalities. Sparse representation-based models, particularly dictionary learning (DL), play an important role in image modeling. Traditionally, DL transforms higher-order tensors into vectors and aggregates them into matrices, disregarding the multi-dimensional inherent structure of data. To overcome this problem, tensor-based DL approaches have been developed. In this study, we propose a tensor-based DL algorithm named CircWaveDL for OCT classification where both the training data and the dictionary are higher-order tensors. Instead of random initialization of the dictionary, we suggested initializing it with CircWave atoms, which has previously demonstrated its effectiveness in OCT classification. This algorithm employs CANDECOMP/PARAFAC (CP) decomposition to factorize each tensor into lower dimensions. Subsequently, we learn a sub-dictionary for each class using the training tensor of that class. A test tensor is reconstructed using each sub-dictionary individually and every test B-scan is assigned to the class with the minimal residual error. To assess the generalizability of the model, we have tested it on three different databases. Furthermore, we introduce a new heatmap generation approach based on averaging the most significant atoms of the learned sub-dictionaries, demonstrating that selecting an appropriate sub-dictionary for test B-scan restoration can lead to better reconstructions, emphasizing distinctive features of different classes. CircWaveDL demonstrates a high level of generalizability according to external validation conducted on three different databases and it outperforms previous classification methods designed for similar datasets.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

CircWaveDL: Modeling of Optical Coherence Tomography Images Based on a new Supervised Tensor-based Dictionary Learning for Classification of Macular abnormalities

Arian,

Vard,

Kafieh

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

A Multichannel Intraluminal Impedance Gastroesophageal Reflux Characterization Algorithm Based On Sparse Representation

Cited by 1 publication

References 38 publications

CircWaveDL: Modeling of Optical Coherence Tomography Images Based on a new Supervised Tensor-based Dictionary Learning for Classification of Macular abnormalities

CircWaveDL: Modeling of Optical Coherence Tomography Images Based on a new Supervised Tensor-based Dictionary Learning for Classification of Macular abnormalities

Contact Info

Product

Resources

About