Left Atrial Ejection Fraction Estimation Using SEGANet for Fully Automated Segmentation of CINE MRI

Lourenço, Ana P.; Kerfoot, Eric; Dibblin, Connor; Alskaf, Ebraham; Anjari, Mustafa; Bharath, Anil A.; King, Andrew P.; Chubb, Henry; Correia, Teresa; Varela, Marta

doi:10.1007/978-3-030-68107-4_14

Cited by 10 publications

(10 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, a large number of ML-based methods have been proposed to automatically perform CMR image segmentation and analysis, thereby significantly reducing the time required for CMR image assessment (337). Considerable efforts have been directed toward cine imaging, as it is considered the gold standard for the assessment of cardiac chamber volumes and function (338)(339)(340)(341). In this case, DL-based methods automatically segment the myocardium and cardiac chambers from MRI images, replacing manual approaches that are time-consuming and prone to observer variability, to enable the extraction of quantitative indices, such as LV and RV volumes, mass, and EF.…”

Section: Image Analysismentioning

confidence: 99%

Cardiac MR: From Theory to Practice

Ismail

Strugnell

Coletti

et al. 2022

Front. Cardiovasc. Med.

View full text Add to dashboard Cite

Cardiovascular disease (CVD) is the leading single cause of morbidity and mortality, causing over 17. 9 million deaths worldwide per year with associated costs of over $800 billion. Improving prevention, diagnosis, and treatment of CVD is therefore a global priority. Cardiovascular magnetic resonance (CMR) has emerged as a clinically important technique for the assessment of cardiovascular anatomy, function, perfusion, and viability. However, diversity and complexity of imaging, reconstruction and analysis methods pose some limitations to the widespread use of CMR. Especially in view of recent developments in the field of machine learning that provide novel solutions to address existing problems, it is necessary to bridge the gap between the clinical and scientific communities. This review covers five essential aspects of CMR to provide a comprehensive overview ranging from CVDs to CMR pulse sequence design, acquisition protocols, motion handling, image reconstruction and quantitative analysis of the obtained data. (1) The basic MR physics of CMR is introduced. Basic pulse sequence building blocks that are commonly used in CMR imaging are presented. Sequences containing these building blocks are formed for parametric mapping and functional imaging techniques. Commonly perceived artifacts and potential countermeasures are discussed for these methods. (2) CMR methods for identifying CVDs are illustrated. Basic anatomy and functional processes are described to understand the cardiac pathologies and how they can be captured by CMR imaging. (3) The planning and conduct of a complete CMR exam which is targeted for the respective pathology is shown. Building blocks are illustrated to create an efficient and patient-centered workflow. Further strategies to cope with challenging patients are discussed. (4) Imaging acceleration and reconstruction techniques are presented that enable acquisition of spatial, temporal, and parametric dynamics of the cardiac cycle. The handling of respiratory and cardiac motion strategies as well as their integration into the reconstruction processes is showcased. (5) Recent advances on deep learning-based reconstructions for this purpose are summarized. Furthermore, an overview of novel deep learning image segmentation and analysis methods is provided with a focus on automatic, fast and reliable extraction of biomarkers and parameters of clinical relevance.

show abstract

Section: Image Analysismentioning

confidence: 99%

Cardiac MR: From Theory to Practice

Ismail

Strugnell

Coletti

et al. 2022

Front. Cardiovasc. Med.

View full text Add to dashboard Cite

show abstract

“…For performance comparison purposes, we trained the U-Net [12], with two dropout layers with a probability of 0.5 to avoid over-fitting and SEGANet, [16] using almost the same training settings as in LA-Net, where we also adjusted weight decay to avoid any over-fitting. We also used transfer learning for both networks when training them with the bSSFP dataset.…”

Section: Experimental Settingsmentioning

confidence: 99%

“…structures such as the left ventricle, there has been comparably less interest in the LA for both clinical and data availability reasons [8]- [10]. Many studies on the segmentation of LA trained the U-Net [12] with dedicated modifications to improve its performance for the LA [13]- [16].…”

mentioning

confidence: 99%

“…The most commonly used protocols for the evaluation of the LA are balanced steady state free precession (bSSFP, also known as trueFISP) and late gadolinium enhanced (LGE) MRI (see Section II-E). When cardiac image analysis literature is examined, one can see a rough grouping of methods targeting a particular MR sequence, either LGE [13], [14], [14], [18], [19] or bSSFP [15], [16], [20]. As far as we are aware, no single algorithm or network has been evaluated on both MR sequences to date.…”

mentioning

confidence: 99%

See 1 more Smart Citation

LA-Net: A Multi-Task Deep Network for the Segmentation of the Left Atrium

Uslu

Varela

Boniface

et al. 2022

IEEE Trans. Med. Imaging

Self Cite

View full text Add to dashboard Cite

Although atrial fibrillation (AF) is the most common sustained atrial arrhythmia, treatment success for this condition remains suboptimal. Information from magnetic resonance imaging (MRI) has the potential to improve treatment efficacy, but there are currently few automatic tools for the segmentation of the atria in MR images. In the study, we propose a LA-Net, a multi-task network optimised to simultaneously generate left atrial segmentation and edge masks from MRI. LA-Net includes cross attention modules (CAMs) and enhanced decoder modules (EDMs) to purposefully select the most meaningful edge information for segmentation and smoothly incorporate it into segmentation masks at multiple-scales. We evaluate the performance of LA-Net on two MR sequences: late gadolinium enhanced (LGE) atrial MRI and atrial short axis balanced steady state free precession (bSSFP) MRI. LA-Net gives Hausdorff distances of 12.43 mm and Dice scores of 0.92 on the LGE (STACOM 2018) dataset and Hausdorff distances of 17.41 mm and Dice scores of 0.90 on the bSSFP (inhouse) dataset without any post-processing, surpassing previously proposed segmentation networks, including U-Net and SEGANet. Our method allows automatic extraction of information about the LA from MR images, which can

show abstract

“…A two-step approach based on NNs is proposed to automatically segment DE-CMR images into 3 classes (LV, healthy myocardium and LGE uptake area) and extract their volumes as additional inputs to Clinic-NET+ and DOC-NET+. These NNs are based on the 2D U-Net architecture [8,14] and were trained separately on the EMIDEC dataset. The first NN was trained with the Dice loss function to identify the LV centre by segmenting the LV blood pool region and calculating the LV centroid coordinates.…”

Section: De-cmr Segmentationmentioning

confidence: 99%

Automatic Myocardial Disease Prediction From Delayed-Enhancement Cardiac MRI and Clinical Information

Lourenço¹,

Kerfoot²,

Grigorescu³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

Delayed-enhancement cardiac magnetic resonance (DE-CMR) provides important diagnostic and prognostic information on myocardial viability. The presence and extent of late gadolinium enhancement (LGE) in DE-CMR is negatively associated with the probability of improvement in left ventricular function after revascularization. Moreover, LGE findings can support the diagnosis of several other cardiomyopathies, but its absence does not rule them out, making disease classification by visual assessment difficult. In this work, we propose deep learning neural networks that can automatically predict myocardial disease from patient clinical information and DE-CMR. All the proposed networks achieve very good classification accuracy (>85%). Including information from DE-CMR (directly as images or as metadata following DE-CMR segmentation) is valuable in this classification task, improving the accuracy to 95-100%.

show abstract

Left Atrial Ejection Fraction Estimation Using SEGANet for Fully Automated Segmentation of CINE MRI

Cited by 10 publications

References 13 publications

Cardiac MR: From Theory to Practice

Cardiac MR: From Theory to Practice

LA-Net: A Multi-Task Deep Network for the Segmentation of the Left Atrium

Automatic Myocardial Disease Prediction From Delayed-Enhancement Cardiac MRI and Clinical Information

Contact Info

Product

Resources

About