2024
Assessment of valve regurgitation severity via contrastive learning and multi-view video integration
Kim S, Ren H, Charton J, Hu J, Gonzalez C, Khambhati J, Cheng J, DeFrancesco J, Waheed A, Marciniak S, Moura F, Cardoso R, Lima B, McKinney S, Picard M, Li X, Li Q. Assessment of valve regurgitation severity via contrastive learning and multi-view video integration. Physics In Medicine And Biology 2024, 69: 045020. PMID: 38271727, DOI: 10.1088/1361-6560/ad22a4.Peer-Reviewed Original ResearchConceptsRepresentation learningContrastive learningMulti-view video processingState-of-the-art methodsContrastive learning frameworkState-of-the-artIn-house datasetVideo processingContrastive networkEmbedding spaceImage inputLearning frameworkVideo integrationIntricate taskLoss termMulti-modal clinical dataLearningRepresentationLabor-intensiveAccuracyEfficient methodVideo seriesEmbeddingDatasetNetwork
2023
Multi-task Learning for Hierarchically-Structured Images: Study on Echocardiogram View Classification
Charton J, Ren H, Kim S, Gonzalez C, Khambhati J, Cheng J, DeFrancesco J, Waheed A, Marciniak S, Moura F, Cardoso R, Lima B, Picard M, Li X, Li Q. Multi-task Learning for Hierarchically-Structured Images: Study on Echocardiogram View Classification. Lecture Notes In Computer Science 2023, 14337: 185-194. DOI: 10.1007/978-3-031-44521-7_18.Peer-Reviewed Original ResearchDeep learningMulti-task learning schemeVideo classification tasksMulti-task learningImage classification scenariosResidual neural networkData labelsClassification taskLearning schemeAblation studiesClassification scenariosNeural networkTree structureMedical imagesImage-processing techniquesSuperior performanceVideo viewingMonitoring cardiovascular diseasesTraining methodsHierarchically-structuredImage acquisitionLearningHierarchical structureTaskClassification