Matches in SemOpenAlex for { <https://semopenalex.org/work/W3207708956> ?p ?o ?g. }
Showing items 1 to 68 of
68
with 100 items per page.
- W3207708956 abstract "Abstract Introduction Carotid atherosclerosis is a major risk factor for ischaemic stroke, a leading cause of death. Carotid CT angiography (CTA) is commonly performed following a stroke or transient ischaemic attack (TIA) to help guide patient management in secondary prevention of stroke. Deep learning algorithms can help extract greater information from scans. Method The dataset comprised CTA scans from 40 culprit and 40 non-culprit carotid arteries of patients with recent stroke/TIA, and 40 carotid arteries of asymptomatic patients without previous stroke/TIA. A 3D convolutional neural network was trained to classify carotid artery type. Each input comprised 14 axial CTA carotid patches (centred around the carotid artery) concatenated together to form a 3D volume (capturing ∼3cm of artery). 75% of the dataset was used for training and 25% for internal validation. Following training, computer vision operations were applied to input images to assess their impact on the model’s classification decisions. Results The model achieved 100% accuracy on the training set and 67% on the internal validation set. However, after subjecting input images to image operations, vulnerabilities in the deep learning model were revealed, even when using input images from the training set. For example, using a Gaussian blur filter with sigma 1.0 was sufficient to change classification decisions, as was horizontally flipping the image. Conclusions Deep learning has exceptional capabilities for learning, however the risk with such high-capacity models is failure to learn relevant features from the data. Stress testing provides a viable method to further evaluate deep learning models before clinical deployment." @default.
- W3207708956 created "2021-10-25" @default.
- W3207708956 creator A5045025159 @default.
- W3207708956 creator A5066811192 @default.
- W3207708956 creator A5071484052 @default.
- W3207708956 creator A5073596524 @default.
- W3207708956 creator A5084611369 @default.
- W3207708956 date "2021-09-01" @default.
- W3207708956 modified "2023-09-23" @default.
- W3207708956 title "875 Using Stress Testing to Identify Vulnerabilities in Artificial Intelligence Models for the Identification of Culprit Carotid Lesions in Cerebrovascular Events" @default.
- W3207708956 doi "https://doi.org/10.1093/bjs/znab259.1123" @default.
- W3207708956 hasPublicationYear "2021" @default.
- W3207708956 type Work @default.
- W3207708956 sameAs 3207708956 @default.
- W3207708956 citedByCount "0" @default.
- W3207708956 crossrefType "journal-article" @default.
- W3207708956 hasAuthorship W3207708956A5045025159 @default.
- W3207708956 hasAuthorship W3207708956A5066811192 @default.
- W3207708956 hasAuthorship W3207708956A5071484052 @default.
- W3207708956 hasAuthorship W3207708956A5073596524 @default.
- W3207708956 hasAuthorship W3207708956A5084611369 @default.
- W3207708956 hasConcept C108583219 @default.
- W3207708956 hasConcept C126838900 @default.
- W3207708956 hasConcept C127413603 @default.
- W3207708956 hasConcept C154945302 @default.
- W3207708956 hasConcept C164705383 @default.
- W3207708956 hasConcept C2778333808 @default.
- W3207708956 hasConcept C2778826181 @default.
- W3207708956 hasConcept C2780645631 @default.
- W3207708956 hasConcept C2987047532 @default.
- W3207708956 hasConcept C41008148 @default.
- W3207708956 hasConcept C500558357 @default.
- W3207708956 hasConcept C71924100 @default.
- W3207708956 hasConcept C78519656 @default.
- W3207708956 hasConcept C81363708 @default.
- W3207708956 hasConceptScore W3207708956C108583219 @default.
- W3207708956 hasConceptScore W3207708956C126838900 @default.
- W3207708956 hasConceptScore W3207708956C127413603 @default.
- W3207708956 hasConceptScore W3207708956C154945302 @default.
- W3207708956 hasConceptScore W3207708956C164705383 @default.
- W3207708956 hasConceptScore W3207708956C2778333808 @default.
- W3207708956 hasConceptScore W3207708956C2778826181 @default.
- W3207708956 hasConceptScore W3207708956C2780645631 @default.
- W3207708956 hasConceptScore W3207708956C2987047532 @default.
- W3207708956 hasConceptScore W3207708956C41008148 @default.
- W3207708956 hasConceptScore W3207708956C500558357 @default.
- W3207708956 hasConceptScore W3207708956C71924100 @default.
- W3207708956 hasConceptScore W3207708956C78519656 @default.
- W3207708956 hasConceptScore W3207708956C81363708 @default.
- W3207708956 hasIssue "Supplement_6" @default.
- W3207708956 hasLocation W32077089561 @default.
- W3207708956 hasOpenAccess W3207708956 @default.
- W3207708956 hasPrimaryLocation W32077089561 @default.
- W3207708956 hasRelatedWork W2731899572 @default.
- W3207708956 hasRelatedWork W2999805992 @default.
- W3207708956 hasRelatedWork W3011074480 @default.
- W3207708956 hasRelatedWork W3116150086 @default.
- W3207708956 hasRelatedWork W3133861977 @default.
- W3207708956 hasRelatedWork W3192840557 @default.
- W3207708956 hasRelatedWork W4200173597 @default.
- W3207708956 hasRelatedWork W4291897433 @default.
- W3207708956 hasRelatedWork W4312417841 @default.
- W3207708956 hasRelatedWork W4321369474 @default.
- W3207708956 hasVolume "108" @default.
- W3207708956 isParatext "false" @default.
- W3207708956 isRetracted "false" @default.
- W3207708956 magId "3207708956" @default.
- W3207708956 workType "article" @default.