Fully automatic segmentation of right and left ventricle on short-axis cardiac MRI images

Budai, Ádám; Suhai, Ferenc Imre; Csorba, Kristóf; Tóth, Attila; Szabó, L; Vágó, Hajnalka; Merkely, Béla

doi:10.1016/j.compmedimag.2020.101786

Cited by 31 publications

(13 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table 6 illustrates the performance comparison of the proposed ES‐FCN model with the existing methods in terms of Jaccard indices. The ES‐FCN model attains the jaccard index of 0.948 for ACDC dataset which is greater than the existing models such as 34,35 . The ES‐FCN model improves the overall Jaccard indices of 10.6% and 13.8% better than Regression‐based method 34 and Pre‐trained neural network 35 for ACDC dataset.…”

Section: Resultsmentioning

confidence: 84%

“…The ES-FCN model attains the jaccard index of 0.948 for ACDC dataset which is greater than the existing models such as. 34,35 The ES-FCN model improves the overall Jaccard indices of 10.6% and 13.8% better than Regression-based method 34 and Pre-trained neural network 35 for ACDC dataset. The proposed model obtains the jaccard index of 0.947 for LVSV dataset, which is greater than the existing models such as.…”

Section: Evaluation Of Acdc Datasetmentioning

confidence: 94%

See 1 more Smart Citation

Automatic left ventricle segmentation via edge‐shape feature‐based fully convolutional neural network

Gayathri,

Uma Maheswari,

Venkatesh

et al. 2023

Int J Imaging Syst Tech

View full text Add to dashboard Cite

Left ventricle (LV) segmentation is essential to identify the cardiac functions for treating cardiovascular disorders. Cardiovascular magnetic resonance (CMRI) imaging is a non‐invasive technique for diagnosing cardiovascular diseases. CMRI is widely used to assess the functional integrity of the left and right ventricles for detecting changes in myocardial structure. Clinical parameters of the LV are often retrieved from CMRI scans, such as LV volumes, and ejection fraction. Moreover, manually segmenting cardiac diseases and evaluating such functions is time‐consuming and difficult for medical professionals. Deep learning networks require a lot of time, cost, and knowledge. To overcome this issue, a novel Edge and Shape feature‐based Fully Convolutional Neural Network (ES‐FCN) has been proposed for automatic LV segmentation. The ES‐FCN model segments MRI images based on edge maps instead of using gray‐scale images, which accelerates the performance of the FCN. The fuzzy‐based canny edge detection algorithm leverages fuzzy logic to detect structural changes in the LV and generate binary images. Additionally, binary‐valued kernels are used for convolution operations, where the binary values are influenced by biases derived from edge map shape descriptors. In ES‐FCN, bias values are learned from ground truth segmentation. The proposed ES‐FCN model achieves the Jaccard indices of 0.9484 ± 0.0188 for the ACDC dataset and 0.9476 ± 0.0237 for the LVSC dataset, and the dice index of 0.9319 ± 0.0188 for the ACDC dataset and 0.9314 ± 0.0237 for LVSC dataset respectively. The experimental results also reveal that the proposed ES‐FCN model is faster and requires minimal resources compared to state‐of‐the‐art models.

show abstract

Section: Resultsmentioning

confidence: 84%

Section: Evaluation Of Acdc Datasetmentioning

confidence: 94%

Automatic left ventricle segmentation via edge‐shape feature‐based fully convolutional neural network

Gayathri,

Uma Maheswari,

Venkatesh

et al. 2023

Int J Imaging Syst Tech

View full text Add to dashboard Cite

show abstract

“…While it is unlear if this could be related to adding new sequences, overall this data suggests that experience may not decrease scan times or interpretation times. The survey’s results also highlight the need to develop faster scanning techniques and automated image analysis to maximize efficiency [ 23 – 25 ].…”

Section: Discussionmentioning

confidence: 99%

Survey of centers performing cardiovascular magnetic resonance in pediatric and congenital heart disease: a report of the Society for Cardiovascular Magnetic Resonance

Buddhe

Soriano

Powell

2022

Journal of Cardiovascular Magnetic Resonance

View full text Add to dashboard Cite

Background There are few data on practice patterns and trends for cardiovascular magnetic resonance (CMR) in pediatric and congenital heart disease. The Society for Cardiovascular Magnetic Resonance (SCMR) sought to address this deficiency by performing an international survey of CMR centers. Methods Surveys consisting of 31 (2014) and 33 (2018) items were designed to collect data on the use of CMR for the evaluation of pediatric and congenital heart disease patients. They were sent to all SCMR members in 2014 and 2018. One response per center was collected. Results There were 93 centers that responded in 2014 and 83 in 2018. The results that follow show data from 2014 and 2018 separated by a dash. The median annual number of pediatric/congenital CMR cases per center was 183–209. The median number of scanners for CMR was 2–2 (range, 1–8) with 58–63% using only 1.5T scanners and 4–4% using only 3T scanners. The mean number of attending/staff reading CMRs was 3.7–2.6; among them, 52–61% were pediatric or adult cardiologists and 47–38% were pediatric or adult radiologists. The median annual case volume per attending was 54–86. The median number of technologists per center doing CMRs was 4–5. The median scanner time allocated for a non–sedated examination was 75–75 min (range, 45–120). Among the 21 centers responding to both surveys, the mean annual case volume increased from 320 in 2014 to 445 in 2018; 17 (81%) of the centers had an increase in annual case volume. For this subgroup, the median attending/staff per center was 4 in both 2014 and 2018. The median scanner time allotted per study was unchanged at 90 min. The mean time for an attending/staff physician to perform a typical CMR examination including reporting was 143–141 min. Conclusion These survey data provide a novel comprehensive view of CMR practice in pediatric and congenital heart disease. This information is useful for internal benchmarking, resource allocation, addressing practice variation, quality improvement initiatives, and identifying unmet needs.

show abstract

“…Critically, DL-based segmentation tools are often trained and tested on images from single clinical centers, using one vendor with a well-defined protocol resulting in homogenous datasets ( 59 , 60 ). Furthermore, CMR protocols across prominent multi-center cohort studies are also standardized, prohibiting wider generalizability ( 5 , 61 , 62 ). A notable effort to develop segmentation tools on more heterogeneous datasets to promote robust AI tool development is the Multi-Center, Multi-Vendor and Multi-Disease Cardiac Segmentation (M&Ms) Challenge ( 63 ).…”

Section: Unintended Consequences Of Ai Applications In Cardiovascular...mentioning

confidence: 99%

Clinician's guide to trustworthy and responsible artificial intelligence in cardiovascular imaging

Szabó

Raisi‐Estabragh²,

Salih³

et al. 2022

Front. Cardiovasc. Med.

Self Cite

View full text Add to dashboard Cite

A growing number of artificial intelligence (AI)-based systems are being proposed and developed in cardiology, driven by the increasing need to deal with the vast amount of clinical and imaging data with the ultimate aim of advancing patient care, diagnosis and prognostication. However, there is a critical gap between the development and clinical deployment of AI tools. A key consideration for implementing AI tools into real-life clinical practice is their “trustworthiness” by end-users. Namely, we must ensure that AI systems can be trusted and adopted by all parties involved, including clinicians and patients. Here we provide a summary of the concepts involved in developing a “trustworthy AI system.” We describe the main risks of AI applications and potential mitigation techniques for the wider application of these promising techniques in the context of cardiovascular imaging. Finally, we show why trustworthy AI concepts are important governing forces of AI development.

show abstract

Fully automatic segmentation of right and left ventricle on short-axis cardiac MRI images

Cited by 31 publications

References 17 publications

Automatic left ventricle segmentation via edge‐shape feature‐based fully convolutional neural network

Automatic left ventricle segmentation via edge‐shape feature‐based fully convolutional neural network

Survey of centers performing cardiovascular magnetic resonance in pediatric and congenital heart disease: a report of the Society for Cardiovascular Magnetic Resonance

Clinician's guide to trustworthy and responsible artificial intelligence in cardiovascular imaging

Contact Info

Product

Resources

About