Attention graph convolutional nets for esophageal contraction pattern recognition in high-resolution manometries

Wang, Zheng; Lu, Yan; Dai, Yuzhuo; Lu, Fanggen; Zhang, Jie; Hou, Muzhou; Liu, Xiaowei

doi:10.1016/j.bspc.2021.102734

Cited by 3 publications

(3 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…More recently, we developed an unsupervised machine learning model on esophageal swallow-level data for feature extraction to identify contraction patterns and separate swallow types. Wang et al introduced a novel contractile vigor propagation graph attention network (CVP-GAT) model to recognize esophageal contraction patterns from high-resolution manometry images [12]. With extracted feature vectors of contractive vigor, their model achieved good accuracy in separating related esophageal motor disorder in three classes at the swallow level (normal, weak and failed).…”

Section: Introductionmentioning

confidence: 99%

A multi-stage machine learning model for diagnosis of esophageal manometry

Kou

Carlson

Baumann

et al. 2022

Artificial Intelligence in Medicine

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

A multi-stage machine learning model for diagnosis of esophageal manometry

Kou

Carlson

Baumann

et al. 2022

Artificial Intelligence in Medicine

View full text Add to dashboard Cite

“…The most crucial drawback of systems relying on deep neural networks is their need for a considerable amount of data for training, which was mentioned by [13]. While esophageal motility problems are not very common, their relatively high diversity means that data are scarce for each disorder.…”

Section: Discussionmentioning

confidence: 99%

“…For instance, Wang et al located swallows using a 2D convolutional neural network and then extracted features from a bidirectional convolutional LSTM using a dense network to classify data into three ranks of disorder severity [12]. In [13], the same rating scheme was applied to the state of contractile vigor of peristalsis utilizing the propagation model of vigor and an graph attention network.…”

Section: Introductionmentioning

confidence: 99%

A fuzzy-based framework for diagnosing esophageal mobility disorder using high-resolution manometry

Rafieivand

Moradi

Sanat

et al. 2023

Journal of Biomedical Informatics

View full text Add to dashboard Cite

A fuzzy-based framework for diagnosing esophageal motility disorder using high-resolution manometry

Rafieivand

Moradi

Sanat

et al. 2022

Preprint

View full text Add to dashboard Cite

In recent years, the high-resolution manometry (HRM) technique has been increasingly used to study esophageal and colonic pressurization and has become a standard routine for discovering mobility disorders. In addition to evolving guidelines for the interpretation of HRM like Chicago, some complexities, such as the dependency of normative reference values on the recording device and other external variables, still remain for medical professions. In this study, a decision support framework is developed to aid the diagnosis of esophageal motility disorders based on HRM data. To abstract HRM data, a new graph-based representation method is introduced that is derived from the spatio-temporal dependencies of pressure values of HRM components. Convolutional graph networks are then utilized to embed relation graphs to the features vector. In the decision-making stage, a novel Expert per Class Fuzzy Classifier (EPC-FC) is presented that employs the ensemble structure and contains expertized sub-classifiers for recognizing a specific disorder. Training sub-classifiers using the negative correlation learning method makes the EPC-FC highly generalizable. Meanwhile, separating the sub-classifiers of each class gives flexibility and interpretability to the structure. The classification results show that our system can distinguish motility disorders with an average accuracy of 78.03% for a single swallow and 92.54% for subject-level classification. Our framework outperforms other comparative classifiers such as SVM and AdaBoost. Moreover, compared with the other studies, the suggested framework has an outstanding performance considering that it imposes no limits on the type of classes or HRM data.

show abstract

Attention graph convolutional nets for esophageal contraction pattern recognition in high-resolution manometries

Cited by 3 publications

References 24 publications

A multi-stage machine learning model for diagnosis of esophageal manometry

A multi-stage machine learning model for diagnosis of esophageal manometry

A fuzzy-based framework for diagnosing esophageal mobility disorder using high-resolution manometry

A fuzzy-based framework for diagnosing esophageal motility disorder using high-resolution manometry

Contact Info

Product

Resources

About