FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3177062007> ?p ?o ?g. }

• W3177062007 abstract "Background: Colonoscopy remains the gold-standard screening for colorectal cancer. However, significant miss rates for polyps have been reported, particularly when there are multiple small adenomas. This presents an opportunity to leverage computer-aided systems to support clinicians and reduce the number of polyps missed. Method: In this work we introduce the Focus U-Net, a novel dual attention-gated deep neural network, which combines efficient spatial and channel-based attention into a single Focus Gate module to encourage selective learning of polyp features. The Focus U-Net further incorporates short-range skip connections and deep supervision. Furthermore, we introduce the Hybrid Focal loss, a new compound loss function based on the Focal loss and Focal Tversky loss, to handle class-imbalanced image segmentation. For our experiments, we selected five public datasets containing images of polyps obtained during optical colonoscopy: CVC-ClinicDB, Kvasir-SEG, CVC-ColonDB, ETIS-Larib PolypDB and EndoScene test set. To evaluate model performance, we use the Dice similarity coefficient (DSC) and Intersection over Union (IoU) metrics. Results: Our model achieves state-of-the-art results for both CVC-ClinicDB and Kvasir-SEG, with a mean DSC of 0.941 and 0.910, respectively. When evaluated on a combination of five public polyp datasets, our model similarly achieves state-of-the-art results with a mean DSC of 0.878 and mean IoU of 0.809, a 14% and 15% improvement over the previous state-of-the-art results of 0.768 and 0.702, respectively. Conclusions: This study shows the potential for deep learning to provide fast and accurate polyp segmentation results for use during colonoscopy. The Focus U-Net may be adapted for future use in newer non-invasive screening and more broadly to other biomedical image segmentation tasks involving class imbalance and requiring efficiency." @default.
• W3177062007 created "2021-07-05" @default.
• W3177062007 creator A5028460960 @default.
• W3177062007 creator A5033880300 @default.
• W3177062007 creator A5062142185 @default.
• W3177062007 creator A5081459146 @default.
• W3177062007 date "2021-05-16" @default.
• W3177062007 modified "2023-09-27" @default.
• W3177062007 title "Focus U-Net: A novel dual attention-gated CNN for polyp segmentation during colonoscopy" @default.
• W3177062007 cites W1533861849 @default.
• W3177062007 cites W1815076433 @default.
• W3177062007 cites W1901129140 @default.
• W3177062007 cites W1903029394 @default.
• W3177062007 cites W1969651471 @default.
• W3177062007 cites W1981444270 @default.
• W3177062007 cites W2021088830 @default.
• W3177062007 cites W2034269173 @default.
• W3177062007 cites W2063979222 @default.
• W3177062007 cites W2082627290 @default.
• W3177062007 cites W2091723114 @default.
• W3177062007 cites W2106904071 @default.
• W3177062007 cites W2110304648 @default.
• W3177062007 cites W2127192465 @default.
• W3177062007 cites W2140753590 @default.
• W3177062007 cites W2146430570 @default.
• W3177062007 cites W2147072570 @default.
• W3177062007 cites W2167597393 @default.
• W3177062007 cites W2302003756 @default.
• W3177062007 cites W2320150036 @default.
• W3177062007 cites W2560023338 @default.
• W3177062007 cites W2560770519 @default.
• W3177062007 cites W2593488799 @default.
• W3177062007 cites W2613041730 @default.
• W3177062007 cites W2624917431 @default.
• W3177062007 cites W2739592885 @default.
• W3177062007 cites W2793703004 @default.
• W3177062007 cites W2884585870 @default.
• W3177062007 cites W2888358068 @default.
• W3177062007 cites W2889646458 @default.
• W3177062007 cites W2899986319 @default.
• W3177062007 cites W2928133111 @default.
• W3177062007 cites W2955058313 @default.
• W3177062007 cites W2962767316 @default.
• W3177062007 cites W2962927567 @default.
• W3177062007 cites W2963351448 @default.
• W3177062007 cites W2963420686 @default.
• W3177062007 cites W2963495494 @default.
• W3177062007 cites W2963606038 @default.
• W3177062007 cites W2963794428 @default.
• W3177062007 cites W2964098128 @default.
• W3177062007 cites W2964189376 @default.
• W3177062007 cites W2964231884 @default.
• W3177062007 cites W2978294504 @default.
• W3177062007 cites W2978457402 @default.
• W3177062007 cites W2980278535 @default.
• W3177062007 cites W2987175876 @default.
• W3177062007 cites W2996290406 @default.
• W3177062007 cites W2999580839 @default.
• W3177062007 cites W3000643822 @default.
• W3177062007 cites W3003166349 @default.
• W3177062007 cites W3034552520 @default.
• W3177062007 cites W3081752372 @default.
• W3177062007 cites W3092344722 @default.
• W3177062007 cites W3100548413 @default.
• W3177062007 cites W3112701542 @default.
• W3177062007 cites W3157328277 @default.
• W3177062007 cites W3160038630 @default.
• W3177062007 cites W3169359689 @default.
• W3177062007 hasPublicationYear "2021" @default.
• W3177062007 type Work @default.
• W3177062007 sameAs 3177062007 @default.
• W3177062007 citedByCount "0" @default.
• W3177062007 crossrefType "posted-content" @default.
• W3177062007 hasAuthorship W3177062007A5028460960 @default.
• W3177062007 hasAuthorship W3177062007A5033880300 @default.
• W3177062007 hasAuthorship W3177062007A5062142185 @default.
• W3177062007 hasAuthorship W3177062007A5081459146 @default.
• W3177062007 hasConcept C103278499 @default.
• W3177062007 hasConcept C108583219 @default.
• W3177062007 hasConcept C115961682 @default.
• W3177062007 hasConcept C120665830 @default.
• W3177062007 hasConcept C121332964 @default.
• W3177062007 hasConcept C121608353 @default.
• W3177062007 hasConcept C124504099 @default.
• W3177062007 hasConcept C126322002 @default.
• W3177062007 hasConcept C153083717 @default.
• W3177062007 hasConcept C153180895 @default.
• W3177062007 hasConcept C154945302 @default.
• W3177062007 hasConcept C163892561 @default.
• W3177062007 hasConcept C169903167 @default.
• W3177062007 hasConcept C192209626 @default.
• W3177062007 hasConcept C2778435480 @default.
• W3177062007 hasConcept C41008148 @default.
• W3177062007 hasConcept C526805850 @default.
• W3177062007 hasConcept C71924100 @default.
• W3177062007 hasConcept C89600930 @default.
• W3177062007 hasConceptScore W3177062007C103278499 @default.
• W3177062007 hasConceptScore W3177062007C108583219 @default.
• W3177062007 hasConceptScore W3177062007C115961682 @default.
• W3177062007 hasConceptScore W3177062007C120665830 @default.

• next