Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313399992> ?p ?o ?g. }
Showing items 1 to 84 of
84
with 100 items per page.
- W4313399992 endingPage "104557" @default.
- W4313399992 startingPage "104557" @default.
- W4313399992 abstract "Echocardiography is a very important medical examination that helps in the computation of critical heart functions. Boundary identification, segmentation and estimation of the volume of key parts of the heart, especially the left ventricle, is a difficult and time-consuming process, even for the most experienced cardiologists. In recent years, research has focused on the automatic segmentation of heart through artificial intelligence techniques and especially with the use of deep learning. Our work is part of this framework. We implemented an ensemble of convolutional neural networks based on the U-net architecture, trained it using a public dataset of cardiac ultrasound images, and combined the outcomes to extract the areas of the left ventricle, myocardium and left atrium. In order to optimize the training process, we have developed a significant data augmentation method based on medical practice. Furthermore, we extended the Dice loss function by imposing additional mandatory anatomical constraints. An ablation study highlights the contribution of each of our proposed modules. The evaluation of our method showed an overall improvement in segmentation accuracy but also in the estimation of clinical metrics. Specifically, using the Dice coefficient for geometric metrics, we achieved for the epicardium a score of 0.96 and 0.955 for the end-diastolic and end-systolic phase respectively. For the clinical metrics of the left ventricle volume, the Pearson correlation coefficient was used where our method gave 0.977, 0.981, 0.897 for the end-diastolic, end-systolic phase and ejection fraction respectively. Scores which up until the writing of this article outperform competitive methods." @default.
- W4313399992 created "2023-01-06" @default.
- W4313399992 creator A5047096552 @default.
- W4313399992 creator A5071354740 @default.
- W4313399992 creator A5076710234 @default.
- W4313399992 date "2023-04-01" @default.
- W4313399992 modified "2023-10-06" @default.
- W4313399992 title "GUDU: Geometrically-constrained Ultrasound Data augmentation in U-Net for echocardiography semantic segmentation" @default.
- W4313399992 cites W2034251649 @default.
- W4313399992 cites W2048309569 @default.
- W4313399992 cites W2119249988 @default.
- W4313399992 cites W2160754664 @default.
- W4313399992 cites W2620296437 @default.
- W4313399992 cites W2905562891 @default.
- W4313399992 cites W2954996726 @default.
- W4313399992 cites W2997310757 @default.
- W4313399992 cites W3036107862 @default.
- W4313399992 cites W3046667076 @default.
- W4313399992 cites W3093273221 @default.
- W4313399992 cites W3093539623 @default.
- W4313399992 cites W3101612813 @default.
- W4313399992 cites W3103215654 @default.
- W4313399992 cites W4249896096 @default.
- W4313399992 doi "https://doi.org/10.1016/j.bspc.2022.104557" @default.
- W4313399992 hasPublicationYear "2023" @default.
- W4313399992 type Work @default.
- W4313399992 citedByCount "4" @default.
- W4313399992 countsByYear W43133999922023 @default.
- W4313399992 crossrefType "journal-article" @default.
- W4313399992 hasAuthorship W4313399992A5047096552 @default.
- W4313399992 hasAuthorship W4313399992A5071354740 @default.
- W4313399992 hasAuthorship W4313399992A5076710234 @default.
- W4313399992 hasConcept C124504099 @default.
- W4313399992 hasConcept C126838900 @default.
- W4313399992 hasConcept C143409427 @default.
- W4313399992 hasConcept C143753070 @default.
- W4313399992 hasConcept C153180895 @default.
- W4313399992 hasConcept C154945302 @default.
- W4313399992 hasConcept C163892561 @default.
- W4313399992 hasConcept C164705383 @default.
- W4313399992 hasConcept C180400216 @default.
- W4313399992 hasConcept C2778198053 @default.
- W4313399992 hasConcept C2778921608 @default.
- W4313399992 hasConcept C41008148 @default.
- W4313399992 hasConcept C71924100 @default.
- W4313399992 hasConcept C78085059 @default.
- W4313399992 hasConcept C81363708 @default.
- W4313399992 hasConcept C89600930 @default.
- W4313399992 hasConceptScore W4313399992C124504099 @default.
- W4313399992 hasConceptScore W4313399992C126838900 @default.
- W4313399992 hasConceptScore W4313399992C143409427 @default.
- W4313399992 hasConceptScore W4313399992C143753070 @default.
- W4313399992 hasConceptScore W4313399992C153180895 @default.
- W4313399992 hasConceptScore W4313399992C154945302 @default.
- W4313399992 hasConceptScore W4313399992C163892561 @default.
- W4313399992 hasConceptScore W4313399992C164705383 @default.
- W4313399992 hasConceptScore W4313399992C180400216 @default.
- W4313399992 hasConceptScore W4313399992C2778198053 @default.
- W4313399992 hasConceptScore W4313399992C2778921608 @default.
- W4313399992 hasConceptScore W4313399992C41008148 @default.
- W4313399992 hasConceptScore W4313399992C71924100 @default.
- W4313399992 hasConceptScore W4313399992C78085059 @default.
- W4313399992 hasConceptScore W4313399992C81363708 @default.
- W4313399992 hasConceptScore W4313399992C89600930 @default.
- W4313399992 hasFunder F4320309480 @default.
- W4313399992 hasLocation W43133999921 @default.
- W4313399992 hasOpenAccess W4313399992 @default.
- W4313399992 hasPrimaryLocation W43133999921 @default.
- W4313399992 hasRelatedWork W2010791679 @default.
- W4313399992 hasRelatedWork W2033207011 @default.
- W4313399992 hasRelatedWork W2362265450 @default.
- W4313399992 hasRelatedWork W2375703560 @default.
- W4313399992 hasRelatedWork W2381206123 @default.
- W4313399992 hasRelatedWork W2410538963 @default.
- W4313399992 hasRelatedWork W2418353436 @default.
- W4313399992 hasRelatedWork W2790347565 @default.
- W4313399992 hasRelatedWork W2912421895 @default.
- W4313399992 hasRelatedWork W4200528772 @default.
- W4313399992 hasVolume "82" @default.
- W4313399992 isParatext "false" @default.
- W4313399992 isRetracted "false" @default.
- W4313399992 workType "article" @default.