Deep Learning for Diagnosis of Chronic Myocardial Infarction on                    Nonenhanced Cardiac Cine MRI

Zhang, Nan; Yang, Guang; Gao, Zhifan; Xu, Chenchu; Zhang, Yanping; Shi, Rui; Keegan, Jennifer; Xu, Lei; Zhang, Heye; Fan, Zhongjie; Firmin, David

doi:10.1148/radiol.2019182304

Cited by 176 publications

(109 citation statements)

References 42 publications

(25 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…They mostly focus on specific view [4] or single frames (i.e., without considering the sequence) [5] or one single vendor and center [6]. As for sequence segmentation, existing methods try to leverage temporal information by using a deformable model combined with the optical flow [7,8] or fine-tuning pretrained CNN dynamically with first frame's label till the last frame [9]. The major downsides of these temporal methods are that they are computational cumbersome and not an end-to-end manner.…”

Section: Gaps Across Views Gaps Across Vendors and Centersmentioning

confidence: 99%

Recurrent Aggregation Learning for Multi-view Echocardiographic Sequences Segmentation

Zhang

Yang

et al. 2019

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Multi-view echocardiographic sequences segmentation is crucial for clinical diagnosis. However, this task is challenging due to limited labeled data, huge noise, and large gaps across views. Here we propose a recurrent aggregation learning method to tackle this challenging task. By pyramid ConvBlocks, multi-level and multi-scale features are extracted efficiently. Hierarchical ConvLSTMs next fuse these features and capture spatial-temporal information in multi-level and multi-scale space. We further introduce a double-branch aggregation mechanism for segmentation and classification which are mutually promoted by deep aggregation of multi-level and multi-scale features. The segmentation branch provides information to guide the classification while the classification branch affords multi-view regularization to refine segmentations and further lessen gaps across views. Our method is built as an end-to-end framework for segmentation and classification. Adequate experiments on our multi-view dataset (9000 labeled images) and the CAMUS dataset (1800 labeled images) corroborate that our method achieves not only superior segmentation and classification accuracy but also prominent temporal stability.

show abstract

Section: Gaps Across Views Gaps Across Vendors and Centersmentioning

confidence: 99%

Recurrent Aggregation Learning for Multi-view Echocardiographic Sequences Segmentation

Zhang

Yang

et al. 2019

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…There are several studies utilizing the AI approach to develop predictive models for MI. [24][25][26][27][28][29][30][31][32] Recently, a group developed an ANN model to predict non-ST elevation myocardial infarction (NSTEMI) patients. 24 The model was trained for several risk attributes such as cardiac risk factor, systolic blood pressure, hemoglobin, corrected QT interval (QTc), PR interval, aspartate aminotransferase, alanine aminotransferase, and cardiac troponin that are independently associated with stable NSTEMI.…”

Section: Ai Techniques and ML Algorithms For Prediction Of Myocardialmentioning

confidence: 99%

Using Artificial Intelligence to Manage Thrombosis Research, Diagnosis, and Clinical Management

Mishra

Ashraf

2019

Semin Thromb Hemost

View full text Add to dashboard Cite

Thrombosis development in either arterial or venous system remains a major cause of death and disability worldwide. This poorly controlled in vivo clotting could result in many severe complications including myocardial infarction, venous thromboembolism, stroke, and cerebral venous thrombosis, to name a few. These conditions are collectively known as thromboembolic disorders (TEDs). Appropriate understanding of TEDs is challenging, as they are multifactorial and involve several and often different risk factors. Hence, it requires a collective effort and data from numerous research studies to fully comprehend molecular mechanisms for prediction, prevention, treatment, and overall management of these conditions. To accomplish this arduous feat, a comprehensive approach is required that can compile thousands of available experimental data and transform these into more applicable and purposeful findings. Thus, large datasets could be utilized to generate models that could be predictive of how an individual would respond when subjected to any kind of additional risk factors or surgery, hospitalization, etc., or in the presence of some susceptible genetic variations. Artificial intelligence-based methods harness the capabilities of computer software to imitate human behaviors such as language translation, visual perception, and, most importantly, decision making. These emerging tools, if appropriately explored, might assist in processing of large data and tackle the complexities of identifying novel or interesting pathways that could otherwise be hidden due to their enormity. This narrative review attempts to compile the applications of various subfields of artificial intelligence and machine learning in the context of thrombosis research to date. It further reflects on the potential of artificial intelligence in transforming enormous research data into translational application in the form of predictive computational models.

show abstract

“…The F is matched with F , which is produced by both the L x and U x . We can further map F to the generated labeled and unlabeled data (L x and U x ) to use the image consistency as the semi-supervised information to learn the T M .We demonstrate the performance of our proposed DCDG for the left atrium segmentation [2, 3] on a LGE-CMRI dataset, which plays an important role in the management of atrial fibrillation and myocardial infarction [4,5].…”

Section: Introductionmentioning

confidence: 99%

Discriminative Consistent Domain Generation for Semi-supervised Learning

Zhang

Zhao

et al. 2019

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Deep learning based task systems normally rely on a large amount of manually labeled training data, which is expensive to obtain and subject to operator variations. Moreover, it does not always hold that the manually labeled data and the unlabeled data are sitting in the same distribution. In this paper, we alleviate these problems by proposing a discriminative consistent domain generation (DCDG) approach to achieve a semi-supervised learning. The discriminative consistent domain is achieved by a double-sided domain adaptation. The double-sided domain adaptation aims to make a fusion of the feature spaces of labeled data and unlabeled data. In this way, we can fit the differences of various distributions between labeled data and unlabeled data. In order to keep the discriminativeness of generated consistent domain for the task learning, we apply an indirect learning for the double-sided domain adaptation. Based on the generated discriminative consistent domain, we can use the unlabeled data to learn the task model along with the labeled data via a consistent image generation. We demonstrate the performance of our proposed DCDG on the late gadolinium enhancement cardiac MRI (LGE-CMRI) images acquired from patients with atrial fibrillation in two clinical centers for the segmentation of the left atrium anatomy (LA) and proximal pulmonary veins (PVs). The experiments show that our semi-supervised approach achieves compelling segmentation results, which can prove the robustness of DCDG for the semi-supervised learning using the unlabeled data along with labeled data acquired from a single center or multicenter studies. * These authors contributed equally to this work.

show abstract

Deep Learning for Diagnosis of Chronic Myocardial Infarction on Nonenhanced Cardiac Cine MRI

Cited by 176 publications

References 42 publications

Recurrent Aggregation Learning for Multi-view Echocardiographic Sequences Segmentation

Recurrent Aggregation Learning for Multi-view Echocardiographic Sequences Segmentation

Using Artificial Intelligence to Manage Thrombosis Research, Diagnosis, and Clinical Management

Discriminative Consistent Domain Generation for Semi-supervised Learning

Contact Info

Product

Resources

About