2025
T‑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 proteaseImproving entity recognition using ensembles of deep learning and fine-tuned large language models: A case study on adverse event extraction from VAERS and social media
Li Y, Viswaroopan D, He W, Li J, Zuo X, Xu H, Tao C. Improving entity recognition using ensembles of deep learning and fine-tuned large language models: A case study on adverse event extraction from VAERS and social media. Journal Of Biomedical Informatics 2025, 163: 104789. PMID: 39923968, DOI: 10.1016/j.jbi.2025.104789.Peer-Reviewed Original ResearchConceptsTraditional deep learning modelsDeep learning modelsRecurrent neural networkLearning modelsEntity recognitionLanguage modelF1 scoreEnsemble of deep learningAdvances of natural language processingEffectiveness of ensemble methodsMicro-averaged F1Bidirectional Encoder RepresentationsExtensive labeled dataNatural language processingFine-tuned modelsBiomedical text miningFeature representationEncoder RepresentationsEvent extractionEntity typesText dataDeep learningSequential dataGPT-2Neural networkImproving the Robustness of Deep-Learning Models in Predicting Hematoma Expansion from Admission Head CT.
Tran A, Karam G, Zeevi D, Qureshi A, Malhotra A, Majidi S, Murthy S, Park S, Kontos D, Falcone G, Sheth K, Payabvash S. Improving the Robustness of Deep-Learning Models in Predicting Hematoma Expansion from Admission Head CT. American Journal Of Neuroradiology 2025, ajnr.a8650. PMID: 39794133, DOI: 10.3174/ajnr.a8650.Peer-Reviewed Original ResearchFast Gradient Sign MethodDeep learning modelsRobustness of deep learning modelsAdversarial attacksAdversarial imagesAdversarial trainingSign MethodModel robustnessDeploying deep learning modelsDeep learning model performanceConvolutional neural networkImprove model robustnessAcute intracerebral hemorrhageHematoma expansionMulti-threshold segmentationReceiver operating characteristicIntracerebral hemorrhageGradient descentType attacksData perturbationNeural networkProjected GradientTraining setAntihypertensive Treatment of Acute Cerebral HemorrhageThreshold-based segmentation
2024
Enhancing Chest X-ray Diagnostics with Neighbor-assisted Multimodal Integration
Xu C, Pan Y, Hu B, Zhang Y, Hong Y, Yang Y. Enhancing Chest X-ray Diagnostics with Neighbor-assisted Multimodal Integration. 2024, 00: 3872-3876. DOI: 10.1109/bibm62325.2024.10822479.Peer-Reviewed Original ResearchField of medical image analysisMultimodal attention networkLeveraging large-scaleMedical image analysisDeep learning modelsDisease classification performanceMultimodal fusionAblation studiesNeighbor informationAttention mechanismTextual informationClassification performanceVisual featuresMultimodal dataMultimodal informationMedical imagesLearning modelsAugmentation techniquesAttention networkData resourcesExperimental resultsX-ray imagesMedical reportsIntegration mechanismsChest X-ray diagnosticsIntegrating Drug Target Information in Deep Learning Models to Predict the Risk of Adverse Events in Patients with Comorbid Post-Traumatic Stress Disorder and Alcohol Use Disorder
Miranda O, Qi X, Brannock M, Whitworth R, Kosten T, Ryan N, Haas G, Kirisci L, Wang L. Integrating Drug Target Information in Deep Learning Models to Predict the Risk of Adverse Events in Patients with Comorbid Post-Traumatic Stress Disorder and Alcohol Use Disorder. Biomedicines 2024, 12: 2772. PMID: 39767679, PMCID: PMC11673068, DOI: 10.3390/biomedicines12122772.Peer-Reviewed Original ResearchPost-traumatic stress disorderAlcohol use disorder patientsAlcohol use disorderComorbid post-traumatic stress disorderUse disorderOpioid use disorderStress disorderSuicidal behaviorPotential therapeutic medicationsDeep learning modelsTarget informationDisordersElectronic medical recordsLearning modelsDepressionPotential therapeutic effectsNeighborhood-level social determinants of healthUniversity of Pittsburgh Medical CenterRisk of adverse eventsNeighborhood-level social determinantsSocial determinants of healthAdverse eventsMedicationAdverse outcomesDeterminants of healthSpatiotemporal Deep Learning-Based Cine Loop Quality Filter for Handheld Point-of-Care Echocardiography
Mukaddim R, Mackay E, Gessert N, Erkamp R, Sethuraman S, Sutton J, Bharat S, Jutras M, Baloescu C, Moore C, Raju B. Spatiotemporal Deep Learning-Based Cine Loop Quality Filter for Handheld Point-of-Care Echocardiography. IEEE Transactions On Ultrasonics Ferroelectrics And Frequency Control 2024, 71: 1577-1587. PMID: 38700961, DOI: 10.1109/tuffc.2024.3396796.Peer-Reviewed Original ResearchOptical flow framesHigh-quality framesLow-quality framesNeural network architectureDeep learning modelsInput framesFrame levelEcho framesNetwork architectureSpatiotemporal deep learning modelCNN modelTemporal informationLV borderLearning modelsTest datasetSpatial informationFlow frameCNNImage qualityPoint-of-careQuantification algorithmHandheldAutomated quantification algorithmImage artifactsImage interpretationA multimodal vision transformer for interpretable fusion of functional and structural neuroimaging data
Bi Y, Abrol A, Fu Z, Calhoun V. A multimodal vision transformer for interpretable fusion of functional and structural neuroimaging data. Human Brain Mapping 2024, 45: e26783. PMID: 39600159, PMCID: PMC11599617, DOI: 10.1002/hbm.26783.Peer-Reviewed Original ResearchConceptsCross-attention mechanismVision transformerDeep learning modelsBrain disordersCharacteristics of schizophreniaDiagnosis of schizophreniaStructural neuroimaging dataNetwork connectivity matrixData fusion approachAttention mapsMultimodal baselinesFunctional network connectivityFuse informationDeep learningICA algorithmFusion approachGrey matter mapsAI algorithmsFunctional network connectivity matricesLeverage multiple sources of informationGray matter imagesLearning modelsMultiple sources of informationBrain imaging modalitiesNetwork connectivityImproved Prediction of Ligand–Protein Binding Affinities by Meta-modeling
Lee H, Emani P, Gerstein M. Improved Prediction of Ligand–Protein Binding Affinities by Meta-modeling. Journal Of Chemical Information And Modeling 2024, 64: 8684-8704. PMID: 39576762, PMCID: PMC11632770, DOI: 10.1021/acs.jcim.4c01116.Peer-Reviewed Original ResearchBinding affinity predictionAffinity predictionMeta-modelMeta-modeling approachLigand-protein binding affinityState-of-the-art deep learning toolsState-of-the-artBinding affinityDeep learning modelsDeep learning toolsMolecular descriptorsInclusion of featuresVirtual screeningBase modelDatabase scalabilityGeneralization capabilityDiverse modeling approachesTraining databaseApplication benchmarksDrug ligandsLearning modelsLigandPhysicochemical propertiesLearning toolsDevelopment effortsAssessing the impact of deep‐learning assistance on the histopathological diagnosis of serous tubal intraepithelial carcinoma (STIC) in fallopian tubes
Bogaerts J, Steenbeek M, Bokhorst J, van Bommel M, Abete L, Addante F, Brinkhuis M, Chrzan A, Cordier F, Devouassoux‐Shisheboran M, Fernández‐Pérez J, Fischer A, Gilks C, Guerriero A, Jaconi M, Kleijn T, Kooreman L, Martin S, Milla J, Narducci N, Ntala C, Parkash V, de Pauw C, Rabban J, Rijstenberg L, Rottscholl R, Staebler A, Van de Vijver K, Zannoni G, van Zanten M, Bart J, Bentz J, Bosse T, Bulten J, Desouki M, Lastra R, Numan T, Schoolmeester J, Schwartz L, Shih I, Soong T, Turashvili G, Vang R, Volchek M, Aliredjo R, Kusters‐Vandevelde H, de Hullu J, Simons M, van der Laak J. Assessing the impact of deep‐learning assistance on the histopathological diagnosis of serous tubal intraepithelial carcinoma (STIC) in fallopian tubes. The Journal Of Pathology Clinical Research 2024, 10: e70006. PMID: 39439213, PMCID: PMC11496567, DOI: 10.1002/2056-4538.70006.Peer-Reviewed Original ResearchConceptsDeep learning modelsSerous tubal intraepithelial carcinomaArtificial intelligenceAI assistanceDiagnosis of serous tubal intraepithelial carcinomaTubal intraepithelial carcinomaReview timeFallopian tubeIntraepithelial carcinomaAI supportHigh-grade serous ovarian carcinomaSerous ovarian carcinomaStandalone performanceAverage sensitivityGroup of pathologistsAccuracyOvarian carcinomaHistopathological diagnosisPathologist performanceMixed-model analysisDiagnostic certaintyCarcinomaDiagnostic settingImproving tabular data extraction in scanned laboratory reports using deep learning models
Li Y, Wei Q, Chen X, Li J, Tao C, Xu H. Improving tabular data extraction in scanned laboratory reports using deep learning models. Journal Of Biomedical Informatics 2024, 159: 104735. PMID: 39393477, DOI: 10.1016/j.jbi.2024.104735.Peer-Reviewed Original ResearchTree edit distanceOptical character recognitionTable recognitionDeep learning modelsAverage recallAverage precisionState-of-the-art deep learning modelsLearning modelsRegion-of-interest detectionState-of-the-artCharacter recognitionDetection evaluationTree editingTabular dataImpressive resultsLab test resultsLaboratory test reportsClinical documentationRecognitionLaboratory reportsHealthcare organizationsClinical data analysisDecision makingClinical decision makingTest reportsA Flow-based Truncated Denoising Diffusion Model for super-resolution Magnetic Resonance Spectroscopic Imaging
Dong S, Cai Z, Hangel G, Bogner W, Widhalm G, Huang Y, Liang Q, You C, Kumaragamage C, Fulbright R, Mahajan A, Karbasi A, Onofrey J, de Graaf R, Duncan J. A Flow-based Truncated Denoising Diffusion Model for super-resolution Magnetic Resonance Spectroscopic Imaging. Medical Image Analysis 2024, 99: 103358. PMID: 39353335, PMCID: PMC11609020, DOI: 10.1016/j.media.2024.103358.Peer-Reviewed Original ResearchDenoising diffusion modelsDeep learning-based super-resolution methodsLearning-based super-resolution methodsMulti-scale super-resolutionGenerative modelSuper-resolution methodsDeep learning modelsHigh-resolution magnetic resonance spectroscopic imagingHigh-quality imagesPost-processing approachSuper-resolutionFlow-based networksLearning modelsLow resolutionTruncation stepLow-resolution dataSharpness adjustmentNetworkSensitivity restrictionsUncertainty estimationDiffusion modelImagesCapabilitySampling processSpectroscopic imagingCommon and unique brain aging patterns between females and males quantified by large‐scale deep learning
Du Y, Yuan Z, Sui J, Calhoun V. Common and unique brain aging patterns between females and males quantified by large‐scale deep learning. Human Brain Mapping 2024, 45: e70005. PMID: 39225381, PMCID: PMC11369911, DOI: 10.1002/hbm.70005.Peer-Reviewed Original ResearchConceptsBrain functional changesFunctional connectivityCognitive controlBrain agingBrain functionPatterns of brain agingResting-state brain functional connectivityBrain functional interactionsBrain functional connectivityHuman brain functionBrain aging patternsGender commonalitiesAge-related changesDeep learningHealthy participantsNormal agingNegative connectionFunctional changesBrainPositive connectionDeep learning modelsFunctional domainsAge effectsFunctional interactionsCross-validation schemeT-cell receptor binding prediction: A machine learning revolution
Weber A, Pélissier A, Martínez M. T-cell receptor binding prediction: A machine learning revolution. ImmunoInformatics 2024, 15: 100040. DOI: 10.1016/j.immuno.2024.100040.Peer-Reviewed Original ResearchProtein language modelsT cell receptorExtract biological insightsUnlabeled protein sequencesProtein sequencesBinding specificityBiological insightsProtein modelsRepertoire dataDeep learning modelsSequenceBlack-box modelsUnsupervised clustering approachDataset biasEvolution of computational modelsLack of generalityLanguage modelImmunizing sequencesMachine learning effortsCompetitive performanceOpaque modelsBiological propertiesLearning modelsClustering approachSupervised modelsFoundational Models for Pathology and Endoscopy Images: Application for Gastric Inflammation
Kerdegari H, Higgins K, Veselkov D, Laponogov I, Polaka I, Coimbra M, Pescino A, Leja M, Dinis-Ribeiro M, Kanonnikoff T, Veselkov K. Foundational Models for Pathology and Endoscopy Images: Application for Gastric Inflammation. Diagnostics 2024, 14: 1912. PMID: 39272697, PMCID: PMC11394237, DOI: 10.3390/diagnostics14171912.Peer-Reviewed Original ResearchUpper gastrointestinal (GI) cancerIntestinal metaplasiaGastric cancerGastrointestinal (GI) cancersImprove patient outcomesAccuracy of endoscopyCancer mortalityGlobal cancer mortalityArtificial intelligencePatient outcomesGC casesChronic inflammationRegular surveillanceGastric inflammationDeep learning modelsClinical practiceIntegration of artificial intelligenceIntegration of multimodal dataLarge-scale dataPathology image analysisEndoscopyCancerEarly detectionMultimodal dataPathologyCT-based multimodal deep learning for non-invasive overall survival prediction in advanced hepatocellular carcinoma patients treated with immunotherapy
Xia Y, Zhou J, Xun X, Zhang J, Wei T, Gao R, Reddy B, Liu C, Kim G, Yu Z. CT-based multimodal deep learning for non-invasive overall survival prediction in advanced hepatocellular carcinoma patients treated with immunotherapy. Insights Into Imaging 2024, 15: 214. PMID: 39186192, PMCID: PMC11347550, DOI: 10.1186/s13244-024-01784-8.Peer-Reviewed Original ResearchConvolutional-recurrent neural networkAdvanced hepatocellular carcinomaSpatial-temporal informationHepatocellular carcinomaCT scanOverall survival predictionRECIST criteriaClinical variablesPatients treated with immunotherapyExtract spatial-temporal informationFollow-up CT imagesPrognostic modelAdvanced HCC patientsRisk group stratificationDeep learning-based modelTest setDisease statusMethodsThis retrospective studyLog-rank testMultimodal deep learningMulti-modal inputsSurvival predictionDeep learning modelsAnalysis of CT scansPatient's disease statusIdentifying EEG Biomarkers of Depression with Novel Explainable Deep Learning Architectures
Ellis C, Sancho M, Miller R, Calhoun V. Identifying EEG Biomarkers of Depression with Novel Explainable Deep Learning Architectures. Communications In Computer And Information Science 2024, 2156: 102-124. DOI: 10.1007/978-3-031-63803-9_6.Peer-Reviewed Original ResearchDeep learning modelsExplainability methodsExplainability analysisConvolutional neural network architectureLearning modelsRaw electroencephalogramNeural network architectureDeep learning architectureMajor depressive disorderLearning architectureNetwork architectureDeep learningModel architectureMultichannel electroencephalogramTraining approachArchitectureBiomarkers of depressionFrequency bandElectroencephalogramResearch contextDepressive disorderElectroencephalogram biomarkerAccuracyRight hemisphereExplainabilityExploring Schizophrenia Classification in fMRI Data: A Common Spatial Patterns(CSP) Approach for Enhanced Feature Extraction and Classification
Esfahani M, Miller R, Calhoun V. Exploring Schizophrenia Classification in fMRI Data: A Common Spatial Patterns(CSP) Approach for Enhanced Feature Extraction and Classification. Annual International Conference Of The IEEE Engineering In Medicine And Biology Society (EMBC) 2024, 00: 1-4. PMID: 40040201, DOI: 10.1109/embc53108.2024.10782387.Peer-Reviewed Original ResearchConceptsImplementation of deep learning modelsNetwork connectivityUnsupervised dimensionality reduction techniquesTime-varying network connectivityEnhanced feature extractionDimensionality reduction techniquesDeep learning modelsMotor imagery tasksFeature extractionElectroencephalogram signalsTransformation of signalsEEG signalsPrincipal component analysisLearning modelsData typesCSP methodApplication of CSPSchizophrenia classificationFMRI datasetsReduction techniquesImagery tasksDatasetCSPDataClassificationLabel Noise-Robust Ensemble Deep Multimodal Framework For Neuroimaging Data
Rokham H, Falakshahi H, Calhoun V. Label Noise-Robust Ensemble Deep Multimodal Framework For Neuroimaging Data. Annual International Conference Of The IEEE Engineering In Medicine And Biology Society (EMBC) 2024, 00: 1-4. PMID: 40039505, DOI: 10.1109/embc53108.2024.10782672.Peer-Reviewed Original ResearchConceptsLabel noiseEffects of label noiseBrain-based markersSelf-report assessmentsLabel noise problemFunctional MRI dataDeep convolutional frameworkDeep learning modelsK-fold cross-validation techniqueAssessment of diagnosisNosological categoriesCross-validation techniqueNeuroimaging dataMental illnessClassification performanceConvolutional frameworkDiagnostic categoriesDiagnostic classificationEnsemble methodsMultimodal frameworkLearning modelsSubsets of dataBagging approachK-foldNeuroimagingMolLM: a unified language model for integrating biomedical text with 2D and 3D molecular representations
Tang X, Tran A, Tan J, Gerstein M. MolLM: a unified language model for integrating biomedical text with 2D and 3D molecular representations. Bioinformatics 2024, 40: i357-i368. PMID: 38940177, PMCID: PMC11256921, DOI: 10.1093/bioinformatics/btae260.Peer-Reviewed Original ResearchConceptsTransformer encoderDownstream tasksLanguage modelBiomedical textSelf-supervised pre-trainingExplicit 3D representationRepresentation improves performanceDeep learning modelsRepresentation of moleculesContrastive learningSupervisory signalExtract embeddingsRepresentation capabilityJoint representationBiomedical domainPre-trainingTextual dataLearning modelsMolecular representationsModel weightsJupyter NotebookStep-by-step guidanceEncodingProperty predictionStructural informationDeep learning survival model predicts outcome after intracerebral hemorrhage from initial CT scan
Chen Y, Rivier C, Mora S, Lopez V, Payabvash S, Sheth K, Harloff A, Falcone G, Rosand J, Mayerhofer E, Anderson C. Deep learning survival model predicts outcome after intracerebral hemorrhage from initial CT scan. European Stroke Journal 2024, 10: 225-235. PMID: 38880882, PMCID: PMC11569453, DOI: 10.1177/23969873241260154.Peer-Reviewed Original ResearchIntracerebral hemorrhage scoreNon-contrast CT scanIntracerebral hemorrhageCT scanFUNC scoreIntracerebral hemorrhage patientsNon-contrast CTFunctional impairmentSevere disabilityDependent living statusLong-term functional impairmentC-indexPrognostic toolFunctional outcomesTreatment decisionsAcute settingClinical implementationRehabilitation strategiesDependent livingPatientsPredicting functional impairmentLong-term care needsPlanning of patient careDeep learning modelsHemorrhage
This site is protected by hCaptcha and its Privacy Policy and Terms of Service apply