FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3196049305> ?p ?o ?g. }

• W3196049305 abstract "Abstract White matter hyperintensities (WMH) are frequently observed on structural neuroimaging of elderly populations and are associated with cognitive decline and increased risk of dementia. Many existing WMH segmentation algorithms produce suboptimal results in populations with vascular lesions or brain atrophy, or require parameter tuning and are computationally expensive. Additionally, most algorithms do not generate a confidence estimate of segmentation quality, limiting their interpretation. MRI-based segmentation methods are often sensitive to acquisition protocols, scanners, noise-level, and image contrast, failing to generalize to other populations and out-of-distribution datasets. Given these concerns, we propose a novel Bayesian 3D Convolutional Neural Network (CNN) with a U-Net architecture that automatically segments WMH, provides uncertainty estimates of the segmentation output for quality control and is robust to changes in acquisition protocols. We also provide a second model to differentiate deep and periventricular WMH. 432 subjects were recruited to train the CNNs from four multi-site imaging studies. A separate test set of 158 subjects was used for evaluation, including an unseen multi-site study. We compared our model to two established state-of-the-art techniques (BIANCA and DeepMedic), highlighting its accuracy and efficiency. Our Bayesian 3D U-Net achieved the highest Dice similarity coefficient of 0.89 ± 0.08 and the lowest modified Hausdorff distance of 2.98 ± 4.40 mm. We further validated our models highlighting their robustness on ‘clinical adversarial cases’ simulating data with low signal-to-noise ratio, low resolution, and different contrast (stemming from MRI sequences with different parameters). Our pipeline and models are available at: https://hypermapp3r.readthedocs.io" @default.
• W3196049305 created "2021-08-30" @default.
• W3196049305 creator A5005209374 @default.
• W3196049305 creator A5009298184 @default.
• W3196049305 creator A5009946184 @default.
• W3196049305 creator A5011579867 @default.
• W3196049305 creator A5015289350 @default.
• W3196049305 creator A5019609065 @default.
• W3196049305 creator A5024226595 @default.
• W3196049305 creator A5040635626 @default.
• W3196049305 creator A5042741911 @default.
• W3196049305 creator A5046826861 @default.
• W3196049305 creator A5047500033 @default.
• W3196049305 creator A5049309607 @default.
• W3196049305 creator A5053189174 @default.
• W3196049305 creator A5054559729 @default.
• W3196049305 creator A5055622232 @default.
• W3196049305 creator A5055859539 @default.
• W3196049305 creator A5056005902 @default.
• W3196049305 creator A5056054317 @default.
• W3196049305 creator A5058795291 @default.
• W3196049305 creator A5059886692 @default.
• W3196049305 creator A5060603330 @default.
• W3196049305 creator A5063711541 @default.
• W3196049305 creator A5064328939 @default.
• W3196049305 creator A5071238538 @default.
• W3196049305 creator A5080621224 @default.
• W3196049305 creator A5085497748 @default.
• W3196049305 creator A5087679982 @default.
• W3196049305 date "2021-08-19" @default.
• W3196049305 modified "2023-10-18" @default.
• W3196049305 title "Deep Bayesian networks for uncertainty estimation and adversarial resistance of white matter hyperintensity segmentation" @default.
• W3196049305 cites W1901129140 @default.
• W3196049305 cites W1980370294 @default.
• W3196049305 cites W1993156157 @default.
• W3196049305 cites W2014513166 @default.
• W3196049305 cites W2018887826 @default.
• W3196049305 cites W2036887984 @default.
• W3196049305 cites W2040609249 @default.
• W3196049305 cites W2047027336 @default.
• W3196049305 cites W2059784307 @default.
• W3196049305 cites W2064590007 @default.
• W3196049305 cites W2083726603 @default.
• W3196049305 cites W2089435027 @default.
• W3196049305 cites W2103328396 @default.
• W3196049305 cites W2110665483 @default.
• W3196049305 cites W2117340355 @default.
• W3196049305 cites W2122477279 @default.
• W3196049305 cites W2136210722 @default.
• W3196049305 cites W2156717321 @default.
• W3196049305 cites W2194775991 @default.
• W3196049305 cites W2301358467 @default.
• W3196049305 cites W2309640507 @default.
• W3196049305 cites W2464708700 @default.
• W3196049305 cites W2563079942 @default.
• W3196049305 cites W2592582590 @default.
• W3196049305 cites W2605649001 @default.
• W3196049305 cites W2770385342 @default.
• W3196049305 cites W2890430415 @default.
• W3196049305 cites W2903804687 @default.
• W3196049305 cites W2919115771 @default.
• W3196049305 cites W2962914239 @default.
• W3196049305 cites W2963076262 @default.
• W3196049305 cites W2963509048 @default.
• W3196049305 cites W2980934859 @default.
• W3196049305 cites W2982140522 @default.
• W3196049305 cites W2984992886 @default.
• W3196049305 cites W2992511782 @default.
• W3196049305 cites W3128830162 @default.
• W3196049305 doi "https://doi.org/10.1101/2021.08.18.456666" @default.
• W3196049305 hasPublicationYear "2021" @default.
• W3196049305 type Work @default.
• W3196049305 sameAs 3196049305 @default.
• W3196049305 citedByCount "3" @default.
• W3196049305 countsByYear W31960493052021 @default.
• W3196049305 countsByYear W31960493052022 @default.
• W3196049305 countsByYear W31960493052023 @default.
• W3196049305 crossrefType "posted-content" @default.
• W3196049305 hasAuthorship W3196049305A5005209374 @default.
• W3196049305 hasAuthorship W3196049305A5009298184 @default.
• W3196049305 hasAuthorship W3196049305A5009946184 @default.
• W3196049305 hasAuthorship W3196049305A5011579867 @default.
• W3196049305 hasAuthorship W3196049305A5015289350 @default.
• W3196049305 hasAuthorship W3196049305A5019609065 @default.
• W3196049305 hasAuthorship W3196049305A5024226595 @default.
• W3196049305 hasAuthorship W3196049305A5040635626 @default.
• W3196049305 hasAuthorship W3196049305A5042741911 @default.
• W3196049305 hasAuthorship W3196049305A5046826861 @default.
• W3196049305 hasAuthorship W3196049305A5047500033 @default.
• W3196049305 hasAuthorship W3196049305A5049309607 @default.
• W3196049305 hasAuthorship W3196049305A5053189174 @default.
• W3196049305 hasAuthorship W3196049305A5054559729 @default.
• W3196049305 hasAuthorship W3196049305A5055622232 @default.
• W3196049305 hasAuthorship W3196049305A5055859539 @default.
• W3196049305 hasAuthorship W3196049305A5056005902 @default.
• W3196049305 hasAuthorship W3196049305A5056054317 @default.
• W3196049305 hasAuthorship W3196049305A5058795291 @default.
• W3196049305 hasAuthorship W3196049305A5059886692 @default.
• W3196049305 hasAuthorship W3196049305A5060603330 @default.
• W3196049305 hasAuthorship W3196049305A5063711541 @default.

• next