Flow Network Based Cardiac Motion Tracking Leveraging Learned Feature Matching
Parajuli N, Lu A, Stendahl J, Zontak M, Boutagy N, Alkhalil I, Eberle M, Lin B, O’Donnell M, Sinusas A, Duncan J. Flow Network Based Cardiac Motion Tracking Leveraging Learned Feature Matching. Lecture Notes In Computer Science 2017, 10434: 279-286. DOI: 10.1007/978-3-319-66185-8_32.Peer-Reviewed Original ResearchCardiac motion trackingNeural networkMotion trackingTedious feature engineeringSiamese neural networkMotion tracking methodFeature engineeringSiamese networkFeature matchingGraph nodesImage patchesSpatiotemporal problemsTracking algorithmTracking methodEdge weightsNetworkLinear programmingConsistent constraintsAdditional important contributionTrackingAlgorithmDatasetProgrammingNodesMatchingLearning-Based Spatiotemporal Regularization and Integration of Tracking Methods for Regional 4D Cardiac Deformation Analysis
Lu A, Zontak M, Parajuli N, Stendahl J, Boutagy N, Eberle M, Alkhalil I, O’Donnell M, Sinusas A, Duncan J. Learning-Based Spatiotemporal Regularization and Integration of Tracking Methods for Regional 4D Cardiac Deformation Analysis. Lecture Notes In Computer Science 2017, 10434: 323-331. DOI: 10.1007/978-3-319-66185-8_37.Peer-Reviewed Original ResearchDeformation analysisTracking methodCardiac motion trackingDeformation-based methodUltrasound imaging propertiesSurface trackingGood overall estimationCardiac deformation analysisMotion trackingSpatiotemporal regularizationInput sourcesDifferent input sourcesTracking resultsTrackingPerceptron networkRegularization procedureMulti-layered perceptron networkMethod