Deep Learning methods for automatic evaluation of delayed enhancement-MRI. The results of the EMIDEC challenge

Lalande, Alain; Chen, Zhihao; Pommier, Thibaut; Decourselle, Thomas; Qayyum, Abdul; Salomon, Michel; Ginhac, Dominique; Skandarani, Youssef; Boucher, Arnaud; Brahim, Khawla; Bruijne, Marleen de; Camarasa, Robin; Correia, Teresa; Xue, Fu‐Shan; Girum, Kibrom B.; Hennemuth, Anja; Huellebrand, Markus; Hussain, Raabid; Ivantsits, Matthias; Ma, Jun; Meyer, Craig H.; Sharma, Rishabh; Jixi, Shi; Tsekos, Nikolaos V.; Varela, M.; Wang, Xiyue; Yang, Sen; Zhang, Hannu; Zhang, Yichi; Zhou, Yuncheng; Zhuang, Xiahai; Couturier, Raphaël; Mériaudeau, Fabrice

doi:10.48550/arxiv.2108.04016

Cited by 3 publications

(3 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When evaluated in the EMIDEC training dataset with ground-truth labels, the Attri-VAE approach provided accuracy results (0.98) equivalent to the best challenge participants reporting their performance on the same dataset (1.0 (Lourenc ¸o et al, 2021), 0.95 (Shi et al, 2021), 0.94 (Ivantsits et al, 2021) and 0.90 (Sharma et al, 2021)). For the testing EMIDEC dataset (Lalande et al, 2021), the best participant method obtained a decreased accuracy (0.82, (Lourenc ¸o et al, 2021;Girum et al, 2021)), increasing to 0.92 for the challenge organizers (Shi et al, 2021). As for the ACDC dataset, which was tested as an external database (i.e., without considering it in training), classification accuracy was substantially reduced (0.59), being worst than results reported by challenge participants (Bernard et al, 2018) (0.96) to classify between the different pathologies (not only between healthy and myocardial infarction).…”

Section: Discussionmentioning

confidence: 99%

Attri-VAE: attribute-based, disentangled and interpretable representations of medical images with variational autoencoders

Çetin¹,

Cámara²,

Ballester³

2022

Preprint

View full text Add to dashboard Cite

Deep learning (DL) methods where interpretability is intrinsically considered as part of the model are required to better understand the relationship of clinical and imaging-based attributes with DL outcomes, thus facilitating their use in reasoning medical decisions. Latent space representations built with variational autoencoders (VAE) do not ensure individual control of data attributes. Attribute-based methods enforcing attribute disentanglement have been proposed in the literature for classical computer vision tasks in benchmark data. In this paper, we propose a VAE approach, the Attri-VAE, that includes an attribute regularization term to associate clinical and medical imaging attributes with different regularized dimensions in the generated latent space, enabling a better disentangled interpretation of the attributes. Furthermore, the generated attention maps explained the attribute encoding in the regularized latent space dimensions. The Attri-VAE approach analyzed healthy and myocardial infarction patients with clinical, cardiac morphology, and radiomics attributes. The proposed model provided an excellent trade-off between reconstruction fidelity, disentanglement, and interpretability, outperforming state-of-the-art VAE approaches according to several quantitative metrics. The resulting latent space allowed the generation of realistic synthetic data in the trajectory between two distinct input samples or along a specific attribute dimension to better interpret changes between different cardiac conditions.

show abstract

Section: Discussionmentioning

confidence: 99%

Attri-VAE: attribute-based, disentangled and interpretable representations of medical images with variational autoencoders

Çetin¹,

Cámara²,

Ballester³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Based on accurate and robust segmentation results, the morphological attributes of physiological and pathological structures can be quantitatively analyzed, so as to provide useful basis for clinicians to diagnose diseases. Recently, deep learning-based methods have shown significant improvements and achieved state-of-the-art performances in many medical image segmentation tasks like cardiac segmentation [3], [4], [5], abdominal segmentation [6], [7], etc. However, the success of most existing deep learningbased methods relies on a large amount of labeled training data to ease the sub-optimal performance caused by over-fitting and ensure reliable generalization performance on test set, while it is hard and expensive to obtain large-amount well-annotated data in the medical imaging domain where only experts can provide reliable annotations.…”

Section: Introductionmentioning

confidence: 99%

Uncertainty-Guided Mutual Consistency Learning for Semi-Supervised Medical Image Segmentation

Zhang¹,

Jiao²,

Liao³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Medical image segmentation is a fundamental and critical step in many clinical approaches. Semi-supervised learning has been widely applied to medical image segmentation tasks since it alleviates the heavy burden of acquiring expertexamined annotations and takes the advantage of unlabeled data which is much easier to acquire. Although consistency learning has been proven to be an effective approach by enforcing an invariance of predictions under different distributions, existing approaches cannot make full use of region-level shape constraint and boundary-level distance information from unlabeled data. In this paper, we propose a novel uncertainty-guided mutual consistency learning framework to effectively exploit unlabeled data by integrating intra-task consistency learning from up-todate predictions for self-ensembling and cross-task consistency learning from task-level regularization to exploit geometric shape information. The framework is guided by the estimated segmentation uncertainty of models to select out relatively certain predictions for consistency learning, so as to effectively exploit more reliable information from unlabeled data. We extensively validate our proposed method on two publicly available benchmark datasets: Left Atrium Segmentation (LA) dataset and Brain Tumor Segmentation (BraTS) dataset. Experimental results demonstrate that our method achieves performance gains by leveraging unlabeled data and outperforms existing semisupervised segmentation methods.

show abstract

“…Comparative study for EMIDEC myocardial segmentation in LGE-MRI (test leaderboard)[62]. Best values are marked in bold font.…”

mentioning

confidence: 99%

An Improved 3D Deep Learning-Based Segmentation of Left Ventricular Myocardial Diseases from Delayed-Enhancement MRI with Inclusion and Classification Prior Information U-Net (ICPIU-Net)

Brahim

Arega

Boucher

et al. 2022

Sensors

Self Cite

View full text Add to dashboard Cite

Accurate segmentation of the myocardial scar may supply relevant advancements in predicting and controlling deadly ventricular arrhythmias in subjects with cardiovascular disease. In this paper, we propose the architecture of inclusion and classification of prior information U-Net (ICPIU-Net) to efficiently segment the left ventricle (LV) myocardium, myocardial infarction (MI), and microvascular-obstructed (MVO) tissues from late gadolinium enhancement magnetic resonance (LGE-MR) images. Our approach was developed using two subnets cascaded to first segment the LV cavity and myocardium. Then, we used inclusion and classification constraint networks to improve the resulting segmentation of the diseased regions within the pre-segmented LV myocardium. This network incorporates the inclusion and classification information of the LGE-MRI to maintain topological constraints of pathological areas. In the testing stage, the outputs of each segmentation network obtained with specific estimated parameters from training were fused using the majority voting technique for the final label prediction of each voxel in the LGE-MR image. The proposed method was validated by comparing its results to manual drawings by experts from 50 LGE-MR images. Importantly, compared to various deep learning-based methods participating in the EMIDEC challenge, the results of our approach have a more significant agreement with manual contouring in segmenting myocardial diseases.

show abstract

Deep Learning methods for automatic evaluation of delayed enhancement-MRI. The results of the EMIDEC challenge

Cited by 3 publications

References 41 publications

Attri-VAE: attribute-based, disentangled and interpretable representations of medical images with variational autoencoders

Attri-VAE: attribute-based, disentangled and interpretable representations of medical images with variational autoencoders

Uncertainty-Guided Mutual Consistency Learning for Semi-Supervised Medical Image Segmentation

An Improved 3D Deep Learning-Based Segmentation of Left Ventricular Myocardial Diseases from Delayed-Enhancement MRI with Inclusion and Classification Prior Information U-Net (ICPIU-Net)

Contact Info

Product

Resources

About