FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387258388> ?p ?o ?g. }

• W4387258388 abstract "Abstract Purpose During the acquisition of MRI data, patient-, sequence-, or hardware-related factors can introduce artefacts that degrade image quality. Four of the most significant tasks for improving MRI image quality have been bias field correction, super-resolution, motion-, and noise correction. Machine learning has achieved outstanding results in improving MR image quality for these tasks individually, yet multi-task methods are rarely explored. Methods In this study, we developed a model to simultaneously correct for all four aforementioned artefacts using multi-task learning. Two different datasets were collected, one consisting of brain scans while the other pelvic scans, which were used to train separate models, implementing their corresponding artefact augmentations. Additionally, we explored a novel loss function that does not only aim to reconstruct the individual pixel values, but also the image gradients, to produce sharper, more realistic results. The difference between the evaluated methods was tested for significance using a Friedman test of equivalence followed by a Nemenyi post-hoc test. Results Our proposed model generally outperformed other commonly-used correction methods for individual artefacts, consistently achieving equal or superior results in at least one of the evaluation metrics. For images with multiple simultaneous artefacts, we show that the performance of using a combination of models, trained to correct individual artefacts depends heavily on the order that they were applied. This is not an issue for our proposed multi-task model. The model trained using our novel convolutional loss function always outperformed the model trained with a mean squared error loss, when evaluated using Visual Information Fidelity, a quality metric connected to perceptual quality. Conclusion We trained two models for multi-task MRI artefact correction of brain, and pelvic scans. We used a novel loss function that significantly improves the image quality of the outputs over using mean squared error. The approach performs well on real world data, and it provides insight into which artefacts it detects and corrects for. Our proposed model and source code were made publicly available." @default.
• W4387258388 created "2023-10-03" @default.
• W4387258388 creator A5017464151 @default.
• W4387258388 creator A5035857033 @default.
• W4387258388 creator A5037383942 @default.
• W4387258388 creator A5048584953 @default.
• W4387258388 creator A5069811142 @default.
• W4387258388 creator A5079971981 @default.
• W4387258388 creator A5092986417 @default.
• W4387258388 date "2023-10-02" @default.
• W4387258388 modified "2023-10-09" @default.
• W4387258388 title "Improving MR image quality with a multi-task model, using convolutional losses" @default.
• W4387258388 cites W1798436414 @default.
• W4387258388 cites W1812490466 @default.
• W4387258388 cites W1977524417 @default.
• W4387258388 cites W2015159529 @default.
• W4387258388 cites W2117340355 @default.
• W4387258388 cites W2127877311 @default.
• W4387258388 cites W2156992893 @default.
• W4387258388 cites W2157848968 @default.
• W4387258388 cites W2158111076 @default.
• W4387258388 cites W2167343814 @default.
• W4387258388 cites W2590628585 @default.
• W4387258388 cites W2782686167 @default.
• W4387258388 cites W2795380527 @default.
• W4387258388 cites W2885398382 @default.
• W4387258388 cites W2888013437 @default.
• W4387258388 cites W2941219400 @default.
• W4387258388 cites W2945147429 @default.
• W4387258388 cites W2962785568 @default.
• W4387258388 cites W2963372104 @default.
• W4387258388 cites W2963682501 @default.
• W4387258388 cites W2973306787 @default.
• W4387258388 cites W3003650457 @default.
• W4387258388 cites W3005867646 @default.
• W4387258388 cites W3013529009 @default.
• W4387258388 cites W3049688872 @default.
• W4387258388 cites W3095034359 @default.
• W4387258388 cites W3105403262 @default.
• W4387258388 cites W3119151570 @default.
• W4387258388 cites W3120602170 @default.
• W4387258388 cites W3141797743 @default.
• W4387258388 cites W3157723514 @default.
• W4387258388 cites W3161810785 @default.
• W4387258388 cites W3196990271 @default.
• W4387258388 cites W3200290869 @default.
• W4387258388 cites W4225328514 @default.
• W4387258388 doi "https://doi.org/10.1186/s12880-023-01109-z" @default.
• W4387258388 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37784039" @default.
• W4387258388 hasPublicationYear "2023" @default.
• W4387258388 type Work @default.
• W4387258388 citedByCount "0" @default.
• W4387258388 crossrefType "journal-article" @default.
• W4387258388 hasAuthorship W4387258388A5017464151 @default.
• W4387258388 hasAuthorship W4387258388A5035857033 @default.
• W4387258388 hasAuthorship W4387258388A5037383942 @default.
• W4387258388 hasAuthorship W4387258388A5048584953 @default.
• W4387258388 hasAuthorship W4387258388A5069811142 @default.
• W4387258388 hasAuthorship W4387258388A5079971981 @default.
• W4387258388 hasAuthorship W4387258388A5092986417 @default.
• W4387258388 hasBestOaLocation W43872583881 @default.
• W4387258388 hasConcept C115961682 @default.
• W4387258388 hasConcept C118615104 @default.
• W4387258388 hasConcept C119857082 @default.
• W4387258388 hasConcept C153180895 @default.
• W4387258388 hasConcept C154945302 @default.
• W4387258388 hasConcept C160633673 @default.
• W4387258388 hasConcept C162324750 @default.
• W4387258388 hasConcept C187736073 @default.
• W4387258388 hasConcept C2780069185 @default.
• W4387258388 hasConcept C2780451532 @default.
• W4387258388 hasConcept C31972630 @default.
• W4387258388 hasConcept C33923547 @default.
• W4387258388 hasConcept C41008148 @default.
• W4387258388 hasConcept C55020928 @default.
• W4387258388 hasConceptScore W4387258388C115961682 @default.
• W4387258388 hasConceptScore W4387258388C118615104 @default.
• W4387258388 hasConceptScore W4387258388C119857082 @default.
• W4387258388 hasConceptScore W4387258388C153180895 @default.
• W4387258388 hasConceptScore W4387258388C154945302 @default.
• W4387258388 hasConceptScore W4387258388C160633673 @default.
• W4387258388 hasConceptScore W4387258388C162324750 @default.
• W4387258388 hasConceptScore W4387258388C187736073 @default.
• W4387258388 hasConceptScore W4387258388C2780069185 @default.
• W4387258388 hasConceptScore W4387258388C2780451532 @default.
• W4387258388 hasConceptScore W4387258388C31972630 @default.
• W4387258388 hasConceptScore W4387258388C33923547 @default.
• W4387258388 hasConceptScore W4387258388C41008148 @default.
• W4387258388 hasConceptScore W4387258388C55020928 @default.
• W4387258388 hasFunder F4320322972 @default.
• W4387258388 hasIssue "1" @default.
• W4387258388 hasLocation W43872583881 @default.
• W4387258388 hasLocation W43872583882 @default.
• W4387258388 hasOpenAccess W4387258388 @default.
• W4387258388 hasPrimaryLocation W43872583881 @default.
• W4387258388 hasRelatedWork W121273120 @default.
• W4387258388 hasRelatedWork W2002009170 @default.
• W4387258388 hasRelatedWork W2034462085 @default.
• W4387258388 hasRelatedWork W2136485282 @default.
• W4387258388 hasRelatedWork W2141888456 @default.

• next