2009
Using Perturbation theory to reduce noise in diffusion tensor fields
Bansal R, Staib LH, Xu D, Laine AF, Liu J, Peterson BS. Using Perturbation theory to reduce noise in diffusion tensor fields. Medical Image Analysis 2009, 13: 580-597. PMID: 19540791, PMCID: PMC2782748, DOI: 10.1016/j.media.2009.05.001.Peer-Reviewed Original ResearchConceptsTensor fieldsDiffusion tensor fieldsPerturbation theoryMarkov random fieldPrior termDifferent spatial directionsRandom fieldsSymmetric tensorsRiemannian distanceSpatial directionsWhite matter fiber bundlesSmoothed fieldsLikelihood termEigenvaluesOriginal fieldEigenvectorsTensorReal-world datasetsDTI datasetsHomogeneous regionsTheoryLow signalNoiseNoise ratioFine structure
2008
Calculation of the confidence intervals for transformation parameters in the registration of medical images
Bansal R, Staib LH, Laine AF, Xu D, Liu J, Posecion LF, Peterson BS. Calculation of the confidence intervals for transformation parameters in the registration of medical images. Medical Image Analysis 2008, 13: 215-233. PMID: 19138877, PMCID: PMC2891652, DOI: 10.1016/j.media.2008.09.002.Peer-Reviewed Original ResearchMeSH KeywordsAlgorithmsArtificial IntelligenceConfidence IntervalsCorpus CallosumData Interpretation, StatisticalHumansImage EnhancementImage Interpretation, Computer-AssistedImaging, Three-DimensionalMagnetic Resonance ImagingPattern Recognition, AutomatedReproducibility of ResultsSensitivity and SpecificitySubtraction TechniqueConceptsSimilarity transformationMultivariate GaussianLeast squares estimationTransformation parametersMathematical frameworkRandom variablesPresence of noiseCovariance matrixLandmark pointsQuantifying errorsSimilarity parameterAmount of misregistrationInherent technological limitationsAmount of noiseGaussianCoordinatesInevitable errorsReal-world datasetsFunctional relationAmount of blurErrorParametersWorld datasetsNoiseConfidence intervals