Automatic left ventricle volume calculation with explainability through a deep learning weak-supervision methodology

Pérez-Pelegrí, Manuel; Monmeneu, José V.; López‐Lereu, María P.; Pérez-Pelegrí, Lucía; Maceira, Alicia M.; Bodı́, Vicente; Moratal, David

doi:10.1016/j.cmpb.2021.106275

Cited by 14 publications

(8 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The average perpendicular distance (APD) is the distance between an automatically segmented contour and the equivalent manually drawn by an expert; this was calculated and averaged across all contour points in the evaluation. A large number indicates that the two outlines did not match closely [33][34][35][36]. The APD is calculated in millimeters using the PixelSpacing DICOM field for spatial resolution.…”

Section: Evaluation Methodsmentioning

confidence: 99%

An Interpretable CNN for the Segmentation of the Left Ventricle in Cardiac MRI by Real-Time Visualization

Li¹,

Yuan²,

Yang³

et al. 2023

Computer Modeling in Engineering &Amp; Sciences

View full text Add to dashboard Cite

The interpretability of deep learning models has emerged as a compelling area in artificial intelligence research. The safety criteria for medical imaging are highly stringent, and models are required for an explanation. However, existing convolutional neural network solutions for left ventricular segmentation are viewed in terms of inputs and outputs. Thus, the interpretability of CNNs has come into the spotlight. Since medical imaging data are limited, many methods to fine-tune medical imaging models that are popular in transfer models have been built using massive public ImageNet datasets by the transfer learning method. Unfortunately, this generates many unreliable parameters and makes it difficult to generate plausible explanations from these models. In this study, we trained from scratch rather than relying on transfer learning, creating a novel interpretable approach for autonomously segmenting the left ventricle with a cardiac MRI. Our enhanced GPU training system implemented interpretable global average pooling for graphics using deep learning. The deep learning tasks were simplified. Simplification included data management, neural network architecture, and training. Our system monitored and analyzed the gradient changes of different layers with dynamic visualizations in real-time and selected the optimal deployment model. Our results demonstrated that the proposed method was feasible and efficient: the Dice coefficient reached 94.48%, and the accuracy reached 99.7%. It was found that no current transfer learning models could perform comparably to the ImageNet transfer learning architectures. This model is lightweight and more convenient to deploy on mobile devices than transfer learning models.

show abstract

Section: Evaluation Methodsmentioning

confidence: 99%

An Interpretable CNN for the Segmentation of the Left Ventricle in Cardiac MRI by Real-Time Visualization

Li¹,

Yuan²,

Yang³

et al. 2023

Computer Modeling in Engineering &Amp; Sciences

View full text Add to dashboard Cite

show abstract

“…Sixteen of these articles relied on well-known XAI methods and applied them to interpret the outcomes of complex DL-based models (more scarcely ML based) having cardiac images from CMR, nuclear medicine, echocardiography, or histopathology as inputs. The aims of these articles were diverse and included the automatic calculation of the LV volume and ejection fraction 28,35 ; the automatic segmentation of cardiac structures as LV/RV, myocardium, interventricular septal, and posterior LV wall 39,51,55 ; and classification approaches. Classification was investigated in the context of cardiac disorders such as mitral valve diseases, 48 myocardial injury, 50 coronary artery disease (CAD), 34,54 cardiomyopathy, 33 congenital heart disease, 40 and HF 32 ; even in newborn 49 for the presence of intracardiac devices (eg, catheters, pacemaker, and defibrillator leads) or motion 18 ; for severe left atrial dilation and LV hypertrophy 43 ; and for CMR image view.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Class activation mapping-based methods are often used when 2-or 3-dimensional images are available, and some extensions have been proposed in cardiac imaging applications. For instance, Pérez-Pelegrıó et al 28 developed a new explainable approach that combines class activation mapping with U-net to automatically estimate the LV volume in end diastole and obtain the result in the form of a segmentation mask without segmentation labels to train the algorithm. Grad-CAM was used in 7 cardiac imaging studies, either for classification 18,34,40,[48][49][50] or segmentation.…”

Section: Literature Reviewmentioning

confidence: 99%

Explainable Artificial Intelligence and Cardiac Imaging: Toward More Interpretable Models

Salih

Galazzo

Gkontra

et al. 2023

Circ: Cardiovascular Imaging

View full text Add to dashboard Cite

Artificial intelligence applications have shown success in different medical and health care domains, and cardiac imaging is no exception. However, some machine learning models, especially deep learning, are considered black box as they do not provide an explanation or rationale for model outcomes. Complexity and vagueness in these models necessitate a transition to explainable artificial intelligence (XAI) methods to ensure that model results are both transparent and understandable to end users. In cardiac imaging studies, there are a limited number of papers that use XAI methodologies. This article provides a comprehensive literature review of state-of-the-art works using XAI methods for cardiac imaging. Moreover, it provides simple and comprehensive guidelines on XAI. Finally, open issues and directions for XAI in cardiac imaging are discussed.

show abstract

“…They measured the effectiveness of the state-of-the-art deep learning models in segmenting the myocardium and ventricles as well as classifying pathologies. Recently, Pérez-Pelegrí et al [35] proposed a deep learning model based on 3D U-Net to estimate LV volume in the end diastole frame. The proposed method provided explanation for obtaining results in the form of segmentation mask without the need of labels for training.…”

Section: Related Workmentioning

confidence: 99%

Convolutional Neural Network Model to Segment Myocardial Infarction from MRI Images

Shaaf

Jamil²,

Ambar³

2023

Int. J. Onl. Eng.

View full text Add to dashboard Cite

Cardiovascular diseases (CVDs) are considered one of the leading causes of death worldwide. Myocardial infarction (MI) is one of the deadliest cardiac diseases that require more consideration. Recently, cardiac magnetic resonance imaging (MRI) has been applied as a standard technique for assessing such diseases. The segmentation of the left ventricle (LV) and myocardium from MRI images is vital in detecting MI disease at its early stages. The automatic segmentation of LV is still challenging due to the complex structures of MRI images, inhomogeneous LV shape and moving organs around the LV, such as the lungs and diaphragm. Thus, this study proposed a convolutional neural network (CNN) model for LV and myocardium segmentation to detect MI. The layers selection and hyper-parameters fine-tuning were applied before the training phase. The model showed robust performance based on the evaluation metrics such as accuracy, sensitivity, specificity, dice score coefficient (DSC), Jaccard index and intersection over union (IOU) with values of 0.86, 0.91, 0.84, 0.81, 0.69 and 0.83, respectively.

show abstract

Automatic left ventricle volume calculation with explainability through a deep learning weak-supervision methodology

Cited by 14 publications

References 16 publications

An Interpretable CNN for the Segmentation of the Left Ventricle in Cardiac MRI by Real-Time Visualization

An Interpretable CNN for the Segmentation of the Left Ventricle in Cardiac MRI by Real-Time Visualization

Explainable Artificial Intelligence and Cardiac Imaging: Toward More Interpretable Models

Convolutional Neural Network Model to Segment Myocardial Infarction from MRI Images

Contact Info

Product

Resources

About