Graph cut segmentation of the right ventricle in cardiac MRI using multi-scale feature learning

Author(s):  
Xuesong Lu ◽  
Xiangan Chen ◽  
Wei Li ◽  
Yuchuan Qiao
Author(s):  
Abdulaziz Ahmed Hashi ◽  
G. V. Ramesh Prasad ◽  
Philip W. Connelly ◽  
Djeven P. Deva ◽  
Michelle M. Nash ◽  
...  

2020 ◽  
Vol 194 ◽  
pp. 102881
Author(s):  
Michael Edwards ◽  
Xianghua Xie ◽  
Robert I. Palmer ◽  
Gary K.L. Tam ◽  
Rob Alcock ◽  
...  

2020 ◽  
Vol 21 (Supplement_1) ◽  
Author(s):  
J Tong ◽  
P Joseph Francis ◽  
E Lee

Abstract Background The presence of an intra-cardiac mass is always a cause for concern, with regards to not only aetiology, but also treatment of complications. We describe a case series of 2 right sided cardiac myxomas, where the first case described an unusual location for tumour occurrence, while the second case provided insights into complications of a cardiac myxoma. Methods The first case involved a 70 year old asymptomatic lady who was referred for an additional heart sound. A transthoracic echocardiogram (TTE) showed a large, mobile 1.5 X 1.2 cm mass, attached to the atrial surface of septal tricuspid valve leaflet, prolapsing in and out of the right sided chambers. This was confirmed on transoesophageal echocardiogram (TEE). Cardiac MRI (magnetic resonance imaging) showed a similar mass attached to the septal tricuspid leaflet with features consistent with a myxoma. She was referred to cardiothoracic surgery, and 2 lobulated tumours arising the septal tricuspid valve and adjacent posterior leaflet were seen. The tumours were resected and a bio-prosthetic tricuspid valve replacement implanted. Histology of the tumours showed myxomatous degeneration of tricuspid valve, consistent with cardiac myxoma. The second case was a 56 year old lady who had dyspnoea, pedal oedema and an elevated jugular venous pulse on examination. A TTE done showed a large 7 X 4 cm mass extending from the right atrium (RA) into the right ventricle (RV). The left ventricular ejection fraction was 35%. Cardiac MRI confirmed the presence of a large mass in the right ventricle that exerted pressure effects on the ventricular septum and RV anterior free wall. Intra-operatively, a large RA mass attached by a stalk to the fossa ovalis was seen. The mass was excised and histology was consistent with cardiac myxoma. A repeat transthoracic echocardiogram done 2 weeks later showed normalisation of the LVEF. See images below for more information. Conclusion While myxomas are the most common benign cardiac tumours, they occur less commonly in the right atrium, and much less so on the tricuspid valve. Clinical manifestations range from being completely asymptomatic, as in the 1st case, to non-specific constitutional symptoms such as fever or general malaise, and to life-threatening complications. These include embolism to the pulmonary circulation, causing sudden death, or to the systemic circulation through an intra-cardiac shunt, causing strokes. This risk is increased if the tumour is large, polypoidal and friable. Large tumours can also cause obstructive symptoms and heart failure. Thus timely diagnosis with multi-modality imaging tools, and definitive treatment with complete resection of the tumour are essential. Continued monitoring for recurrences of the tumour, which can occur in 1-5% of all cases, should be performed as well. Abstract 479 Figure. Right sided cardiac masses


2021 ◽  
Author(s):  
Jesús García Fernández ◽  
Siamak Mehrkanoon

2014 ◽  
Vol 38 (3) ◽  
pp. 190-201 ◽  
Author(s):  
Jordan Ringenberg ◽  
Makarand Deo ◽  
Vijay Devabhaktuni ◽  
Omer Berenfeld ◽  
Pamela Boyers ◽  
...  

2021 ◽  
pp. 107281
Author(s):  
Yueying Li ◽  
Li Liu ◽  
Lei Zhu ◽  
Huaxiang Zhang

Sensors ◽  
2021 ◽  
Vol 21 (15) ◽  
pp. 5125
Author(s):  
Pengcheng Xu ◽  
Zhongyuan Guo ◽  
Lei Liang ◽  
Xiaohang Xu

In the field of surface defect detection, the scale difference of product surface defects is often huge. The existing defect detection methods based on Convolutional Neural Networks (CNNs) are more inclined to express macro and abstract features, and the ability to express local and small defects is insufficient, resulting in an imbalance of feature expression capabilities. In this paper, a Multi-Scale Feature Learning Network (MSF-Net) based on Dual Module Feature (DMF) extractor is proposed. DMF extractor is mainly composed of optimized Concatenated Rectified Linear Units (CReLUs) and optimized Inception feature extraction modules, which increases the diversity of feature receptive fields while reducing the amount of calculation; the feature maps of the middle layer with different sizes of receptive fields are merged to increase the richness of the receptive fields of the last layer of feature maps; the residual shortcut connections, batch normalization layer and average pooling layer are used to replace the fully connected layer to improve training efficiency, and make the multi-scale feature learning ability more balanced at the same time. Two representative multi-scale defect data sets are used for experiments, and the experimental results verify the advancement and effectiveness of the proposed MSF-Net in the detection of surface defects with multi-scale features.


2013 ◽  
Vol 117 (9) ◽  
pp. 1027-1035 ◽  
Author(s):  
D. Grosgeorge ◽  
C. Petitjean ◽  
J.-N. Dacher ◽  
S. Ruan

Sign in / Sign up

Export Citation Format

Share Document