2025
Vision-language foundation model for generalizable nasal disease diagnosis using unlabeled endoscopic records
Liu X, Gong W, Chen X, Li Z, Liu Y, Wang L, Liu Q, Sun X, Liu X, Chen X, Shi Y, Yu H. Vision-language foundation model for generalizable nasal disease diagnosis using unlabeled endoscopic records. Pattern Recognition 2025, 165: 111646. DOI: 10.1016/j.patcog.2025.111646.BooksLabeled dataGeneralization performanceExpert annotationsArtificial intelligencePre-training datasetSuperior generalization performanceState-of-the-artMedical artificial intelligencePerformance of AI modelsNasal endoscopic imagesLearning frameworkAI modelsMultiple imagesSemantic representationDiagnostic tasksFine-tuningTask-specificUniversal representationDatasetExperimental resultsDisease classificationEndoscopic imagesDiagnosis of diseasesAnnotationFoundation modelTexture and noise dual adaptation for infrared image super-resolution
Huang Y, Miyazaki T, Liu X, Dong Y, Omachi S. Texture and noise dual adaptation for infrared image super-resolution. Pattern Recognition 2025, 163: 111449. DOI: 10.1016/j.patcog.2025.111449.Peer-Reviewed Original ResearchTexture detailsAdversarial lossSuper-ResolutionInfrared image super-resolutionVisible imagesImage Super-ResolutionState-of-the-artIR image qualityVisible light imagesAdversarial trainingExtraction branchUpsampling factorsBlurring artifactsImage processingModel adaptationAdaptive approachSpatial domainImage qualityNoiseInnovation frameworkLight imagesNoise transferDual adaptationImagesTexture distributionSingle-Address-Space FaaS with Jord
Li Y, Bhattacharyya A, Kumar M, Bhattacharjee A, Etsion Y, Falsafi B, Kashyap S, Payer M. Single-Address-Space FaaS with Jord. 2025, 694-707. DOI: 10.1145/3695053.3731108.Peer-Reviewed Original ResearchService-level objectivesMemory isolationFunction-as-a-ServiceState-of-the-art systemsState-of-the-artFAAS systemCloud paradigmSoftware developmentMicroservice workloadsFunction dispatchFunctional semanticsPerformance bottleneckExecution timeVirtual memoryCross-functional communicationMicroservicesCo-designExecutionColocating functionsFaaS.MicroVMsOverheadsServerSemanticsThroughputOntology enrichment using a large language model: Applying lexical, semantic, and knowledge network-based similarity for concept placement
Kollapally N, Geller J, Keloth V, He Z, Xu J. Ontology enrichment using a large language model: Applying lexical, semantic, and knowledge network-based similarity for concept placement. Journal Of Biomedical Informatics 2025, 168: 104865. PMID: 40543734, DOI: 10.1016/j.jbi.2025.104865.Peer-Reviewed Original ResearchSemantic triplesSeed ontologyHuman expertsLogical axiomsPubMed abstractsSimilarity search techniquesState-of-the-artReal-world conceptsNetwork-based filterLanguage modelSemantic correctnessText corpusNetwork-based searchSource of textDomain viewSearch techniqueNetwork-based similaritySemMedDBOntology toolsOntologyIdentified conceptsSource of conceptsPipelineDomains of social determinants of healthAxiomsGLAPAL-H: Global, Local, And Parts Aware Learner for Hydrocephalus Infection Diagnosis in Low-Field MRI
Mukherjee S, Templeton K, Tindimwebwa S, Lin P, Sutin J, Yu M, Peterson M, Truwit C, Schiff S, Monga V. GLAPAL-H: Global, Local, And Parts Aware Learner for Hydrocephalus Infection Diagnosis in Low-Field MRI. IEEE Transactions On Biomedical Engineering 2025, PP: 1-14. PMID: 40489263, DOI: 10.1109/tbme.2025.3578541.Peer-Reviewed Original ResearchFeature extractionState-of-the-art alternativesTraining loss functionMulti-task architectureGlobal feature extractionState-of-the-artLocal feature extractionClassification taskCNN branchesSegmentation masksShallow CNNHolistic featuresLocal featuresTraining imageryLoss functionTwo-classThree-classSegmental branchesArchitecture segmentationArchitectureCustomized approachCNNRegularizationQuality issuesPost-infectious hydrocephalusLabel Space-Induced Pseudo Label Refinement for Multi-Source Black-Box Domain Adaptation
Yoo C, Liu X, Xing F, Woo J, Kang J. Label Space-Induced Pseudo Label Refinement for Multi-Source Black-Box Domain Adaptation. IEEE Transactions On Image Processing 2025, 34: 3181-3193. PMID: 40397626, DOI: 10.1109/tip.2025.3570220.Peer-Reviewed Original ResearchUnsupervised domain adaptationApplication programming interfacePseudo-labelsDomain adaptationLabel refinementConventional unsupervised domain adaptationState-of-the-art approachesTarget modelInitial pseudo labelsState-of-the-artMulti-source settingBenchmark datasetsNoisy samplesSource domainTarget domainCompetitive performanceProgramming interfaceTraining frameworkSelf-trainingRefinement phaseSource dataMulti-sourceSource model parametersExperimental resultsRelationship explorationCLIF-Net: Intersection-guided Cross-view Fusion Network for Infection Detection from Cranial Ultrasound
Yu M, Peterson M, Burgoine K, Harbaugh T, Olupot-Olupot P, Gladstone M, Hagmann C, Cowan F, Weeks A, Morton S, Mulondo R, Mbabazi-Kabachelor E, Schiff S, Monga V. CLIF-Net: Intersection-guided Cross-view Fusion Network for Infection Detection from Cranial Ultrasound. IEEE Transactions On Medical Imaging 2025, PP: 1-1. PMID: 40372847, DOI: 10.1109/tmi.2025.3570316.Peer-Reviewed Original ResearchConvolutional neural network branchesNeural network branchesCross-attention moduleState-of-the-artRobust 3D representationsMultiview imageryFusion networkImage-levelFusion layerFusion blockRepresentative featuresExtract featuresLearning frameworkClass probabilitiesSemantic featuresNetwork branchesDetection techniquesEnhanced featuresGeometric relationsImagesEnhanced performanceFeaturesInfection detectionSpatial positionDataset2.5D Multi-view Averaging Diffusion Model for 3D Medical Image Translation: Application to Low-count PET Reconstruction with CT-less Attenuation Correction
Chen T, Hou J, Zhou Y, Xie H, Chen X, Liu Q, Guo X, Xia M, Duncan J, Liu C, Zhou B. 2.5D Multi-view Averaging Diffusion Model for 3D Medical Image Translation: Application to Low-count PET Reconstruction with CT-less Attenuation Correction. IEEE Transactions On Medical Imaging 2025, PP: 1-1. PMID: 40372846, DOI: 10.1109/tmi.2025.3570342.Peer-Reviewed Original ResearchAttenuation correctionLow-dose PETImage-to-image translationStandard-dose PETPositron emission tomographyPET reconstructionOverall radiation doseCT acquisitionState-of-the-art deep learning methodsCNN-based methodsState-of-the-artMedical image translationRadiation doseDeep learning methodsPatient studiesDiffusion modelHigh computation costHuman patient studiesClinical imaging toolBaseline methodsImage translationMulti-viewCNN-basedMultiple viewsGeneration qualityArtificial Intelligence in the Management of Heart Failure
Cheema B, Hourmozdi J, Kline A, Ahmad F, Khera R. Artificial Intelligence in the Management of Heart Failure. Journal Of Cardiac Failure 2025 PMID: 40345521, DOI: 10.1016/j.cardfail.2025.02.020.Peer-Reviewed Original ResearchArtificial intelligenceState-of-the-art algorithmsData privacy concernsState-of-the-artManagement of heart failureAI-based toolsElectronic health recordsAI solutionsMultimodal dataHeart failureHealth recordsIntegration challengesHeart failure syndromeStructural heart diseaseHeart failure treatmentIntelligenceImplementation challengesModel performanceModel governanceAdvanced diseaseFailure syndromeCardiomyopathy diagnosisFailure treatmentRisk factorsHeart diseaseEmerging Image-Guided Navigation Techniques for Cardiovascular Interventions: A Scoping Review
Roshanfar M, Salimi M, Jang S, Sinusas A, Kim J, Mosadegh B. Emerging Image-Guided Navigation Techniques for Cardiovascular Interventions: A Scoping Review. Bioengineering 2025, 12: 488. PMID: 40428106, PMCID: PMC12108902, DOI: 10.3390/bioengineering12050488.Peer-Reviewed Original ResearchState-of-the-artCardiac interventionsNavigation techniquesAugmented reality systemReal-time navigationReality systemArtificial intelligenceImage-guided navigationDecision supportOptical coherence tomographyComplex cardiac interventionsState-of-the-art imaging modalitiesReduced procedure timeConventional fluoroscopyImage-guidedProcedure timeCoherence tomographyProcedural outcomesClinical studiesImproved accuracyElectrophysiological interventionsRadiation exposureReal-time guidanceTraditional techniquesImaging modalitiesTX-Phase: Secure Phasing of Private Genomes in a Trusted Execution Environment
Dokmai N, Zhu K, Sahinalp S, Cho H. TX-Phase: Secure Phasing of Private Genomes in a Trusted Execution Environment. Lecture Notes In Computer Science 2025, 15647: 325-329. DOI: 10.1007/978-3-031-90252-9_32.Peer-Reviewed Original ResearchTrusted Execution EnvironmentExecution environmentData privacy concernsSide-channel leakageState-of-the-artExtract valuable insightsOpen-source softwarePrivate genomic dataFixed-point arithmeticData confidentialityPrivacy constraintsPrivacy concernsSecurity phaseAlgorithmic techniquesServerGenomic dataEnhanced accuracyPractical performanceImputation workflowAnalysis toolsDatasetPhase algorithmAccuracyPrivacyHaplotype phasingOverall Survival Prediction of Brain Tumor Patients with Multimodal MRI using Swin Unetr
Kim G, Xing F, Kong H, Santarnecchi E, Shih H, Bortfeld T, Fakhri G, Liu X, Choi J, Woo J. Overall Survival Prediction of Brain Tumor Patients with Multimodal MRI using Swin Unetr. 2025, 00: 1-5. DOI: 10.1109/isbi60581.2025.10981128.Peer-Reviewed Original ResearchMultimodal magnetic resonance imagingBrain tumor patientsGlioblastoma patient survivalPersonalized treatment plansSurvival prediction performanceMultimodal MRIOverall survival predictionMagnetic resonance imagingHand-crafted featuresState-of-the-artMulti-scale featuresMulti-task frameworkPatient survivalClinical prognosisTumor patientsGlioblastoma patientsSuperior segmentation accuracyTreatment planningResonance imagingSurvival prediction taskPredictive performanceSurvival predictionPrediction taskBRATS datasetSurvivalSpatiotemporal Learning with Context-Aware Video Tubelets for Ultrasound Video Analysis
Li G, Chen L, Hicks B, Schnittke N, Kessler D, Shunp J, Parker M, Baloescu C, Moore C, Gregory C, Gregory K, Raju B, Kruecker J, Chen A. Spatiotemporal Learning with Context-Aware Video Tubelets for Ultrasound Video Analysis. 2025, 00: 1-5. DOI: 10.1109/isbi60581.2025.10981242.Peer-Reviewed Original ResearchGlobal spatial contextState-of-the-art methodsAligned feature mapState-of-the-artUltrasound video analysisReal-time workflowsComplex spatiotemporal informationContext-awarenessFeature mapsObject detectionFive-fold cross-validationSpatiotemporal learningComputational complexityDetection algorithmDetected ROIsSpatial contextUltrasound videosSpatiotemporal featuresSpatiotemporal informationTubeletsVideoVideo analysisClas-sificationReceptive fieldsClassifierCausal Modeling of FMRI Time-Series for Interpretable Autism Spectrum Disorder Classification
Duan P, Dvornek N, Wang J, Staib L, Duncan J. Causal Modeling of FMRI Time-Series for Interpretable Autism Spectrum Disorder Classification. 2025, 00: 1-5. DOI: 10.1109/isbi60581.2025.10980933.Peer-Reviewed Original ResearchFunctional magnetic resonance imagingAutism spectrum disorderState-of-the-art modelsState-of-the-artFMRI time seriesDeep learning classifierDeep learning modelsTime series informationLearning classifiersClassification accuracyNon-linear interactionsMachine learningLeft precuneusRight precuneusABIDE datasetBrain regionsLearning modelsASD populationSpectrum disorderDisorder classificationASD classificationBrain signalsASD biomarkersDevelopmental disordersCorrelation-based modelsPartially characterized topology guides reliable anchor-free scRNA-integration
He C, Filippidis P, Kleinstein S, Guan L. Partially characterized topology guides reliable anchor-free scRNA-integration. Communications Biology 2025, 8: 561. PMID: 40185996, PMCID: PMC11971424, DOI: 10.1038/s42003-025-07988-y.Peer-Reviewed Original ResearchConceptsBatch effectsRare cell typesSingle-cell RNA sequencingCell typesDownstream statistical analysisScRNA-seqBiological insightsRNA sequencingBatch correctionCell phenotypeCellular resolutionBiological signalsState-of-the-art methodsAdaptive lossDomain adaptation lossState-of-the-artDiverse setBatch integrationHeterogeneous cell distributionReconstruction lossSequenceTriplet lossPhenotypeSignalCell distributionT‑ALPHA: A Hierarchical Transformer-Based Deep Neural Network for Protein–Ligand Binding Affinity Prediction with Uncertainty-Aware Self-Learning for Protein-Specific Alignment
Kyro G, Smaldone A, Shee Y, Xu C, Batista V. T‑ALPHA: A Hierarchical Transformer-Based Deep Neural Network for Protein–Ligand Binding Affinity Prediction with Uncertainty-Aware Self-Learning for Protein-Specific Alignment. Journal Of Chemical Information And Modeling 2025, 65: 2395-2415. PMID: 39965912, DOI: 10.1021/acs.jcim.4c02332.Peer-Reviewed Original ResearchConceptsProtein-Ligand Binding Affinity PredictionBinding affinity predictionState-of-the-art performanceTransformer-based deep neural networksMultimodal feature representationAffinity predictionBinding affinity of small moleculesState-of-the-artDeep neural networksDeep learning modelsAffinity of small moleculesSelf-learning methodSARS-CoV-2 main proteasePredicted binding affinitiesFeature representationBinding affinityOn-target potencyNeural networkDrug discovery applicationsTransformation frameworkLearning modelsScoring functionCrystal structureSelf-learningMain proteaseClassifying Unstructured Text in Electronic Health Records for Mental Health Prediction Models: Large Language Model Evaluation Study
Cardamone N, Olfson M, Schmutte T, Ungar L, Liu T, Cullen S, Williams N, Marcus S. Classifying Unstructured Text in Electronic Health Records for Mental Health Prediction Models: Large Language Model Evaluation Study. JMIR Medical Informatics 2025, 13: e65454. PMID: 39864953, PMCID: PMC11884378, DOI: 10.2196/65454.Peer-Reviewed Original ResearchConceptsElectronic health recordsMental health termsHealth termsClinical expertsEmergency departmentHealth recordsPhysical healthMental healthElectronic health record systemsHealth care provider organizationsMental health-related problemsElectronic health record data setsCare provider organizationsMental health cliniciansMental health disordersHealth-related problemsED episodesHealth cliniciansRecords of individualsState-of-the-artDiagnostic codesHealth disordersHospital readmissionProvider organizationsMortality riskBiomedRAG: A retrieval augmented large language model for biomedicine
Li M, Kilicoglu H, Xu H, Zhang R. BiomedRAG: A retrieval augmented large language model for biomedicine. Journal Of Biomedical Informatics 2025, 162: 104769. PMID: 39814274, PMCID: PMC11837810, DOI: 10.1016/j.jbi.2024.104769.Peer-Reviewed Original Research
2024
Identification and Classification of Images in e-Cigarette-Related Content on TikTok: Unsupervised Machine Learning Image Clustering Approach
Lee J, Murthy D, Ouellette R, Anand T, Kong G. Identification and Classification of Images in e-Cigarette-Related Content on TikTok: Unsupervised Machine Learning Image Clustering Approach. Substance Use & Misuse 2024, 60: 677-683. PMID: 40019898, PMCID: PMC11871408, DOI: 10.1080/10826084.2024.2447415.Peer-Reviewed Original ResearchConceptsImage clustering approachImage clustering modelsState-of-the-artE-cigarette-related contentMachine learning approachSocial media dataImage clusteringAnalysis of visual dataSocial media platformsVisual dataLearning approachMedia dataClustering approachMedia platformsImage-based social media platformsQualitative evaluationSocial mediaImage-based analysisImagesTikTokCluster modelUnsupervisedVideoClustersCloudStyle mixup enhanced disentanglement learning for unsupervised domain adaptation in medical image segmentation
Cai Z, Xin J, You C, Shi P, Dong S, Dvornek N, Zheng N, Duncan J. Style mixup enhanced disentanglement learning for unsupervised domain adaptation in medical image segmentation. Medical Image Analysis 2024, 101: 103440. PMID: 39764933, DOI: 10.1016/j.media.2024.103440.Peer-Reviewed Original ResearchConceptsUnsupervised domain adaptationMedical image segmentationDomain-invariant representationsImage segmentationDomain adaptationDisentanglement learningImage translationUnsupervised domain adaptation approachState-of-the-art methodsDomain shift problemDomain-invariant learningState-of-the-artPublic cardiac datasetsDiverse constraintsAdversarial learningConsistency regularizationContrastive learningFeature spaceSemantic consistencyComprehensive experimentsDomain generalizationData diversityShift problemMedical segmentationCardiac datasets
This site is protected by hCaptcha and its Privacy Policy and Terms of Service apply