In neuroimaging the typical analysis approach is a “mass univariate”, where a univariate model is independently fit to each voxel (volume element) in parallel. This approach can be applied to a large range of datasets including: functional MRI, anatomical MRI (with Voxel-Based Morphometry), PET, EEG and MEG data.

The three major software analysis package for neuroimaging: SPM (Statistical Parametric Mapping) and FSL (FMRIB Software Library) and AFNI (AFNI) provide implementations of mass univariate analyses. While there are many commonalities across software packages, there are also software-specific outputs that can be of interest for the end-user.

This document describes the encoding of results of a mass univariate neuroimaging analysis using the NIDM data model ([[!KEATORNI13]]). The goal of this specification is to provide a unified representation of neuroimaging results across analysis software. When a piece of information is only available in a specific software, software-specific extensions are provided.

NIDM-Results is a part of NIDM as depicted in .

NIDM-Results as part of NIDM
NIDM-Results as part of NIDM.

NIDM Family of Documents

This document is part of the NIDM family of documents.

Please Send Comments

This document was published by the NIDASH NIDM Working Group as a Candidate Release. If you wish to make comments regarding this document, please report using the NIDM issue tracker or send them to nidm-users@googlegroups.com. All comments are welcome.

Introduction

NIDM-Results is a NIDM compliant structured representation of the results of a massively univariate neuroimaging study.

NIDM-Results specification formally describes the encoding of the massively univariate neuroimaging results into a NIDM representation. Extensions are provided to describe software-specific objects for SPM (Statistical Parametric Mapping) and FSL (FMRIB Software Library) analysis software.

Structure of this Document

Section 2 provides an overview of NIDM-Results, distinguishing a core set of types and relations, commonly found in provenance, from extended structures catering for more specific uses. It also introduces a modular organization of the data model in components.

Section 3 illustrates how NIDM-Results can be used to express the results of a neuroimaging study in SPM and FSL.

Section 4 provides the definitions (and examples) of NIDM-Results concepts, structured according to three components.

Notational Conventions

The key word "OPTIONAL" in this document is to be interpreted as described in [RFC2119].

Examples throughout this document use the PROV-N Provenance Notation, specified in a separate document [[prov-n]].

Namespaces

The following namespaces prefixes are used throughout this document.

Table 1: Prefix and Namespaces used in this specification
prefixnamespace IRI definition
nidmhttp://www.incf.org/ns/nidash/nidm#The NIDM namespace
provhttp://www.w3.org/ns/prov#The PROV namespace [[prov-dm]]
xsdhttp://www.w3.org/2000/10/XMLSchema#XML Schema Namespace [[XMLSCHEMA11-2]]
(others)(various)All other namespace prefixes are used in examples only.
In particular, IRIs starting with "http://example.com" represent
some application-dependent IRI [[RFC3987]]

Overview

This section introduces neuroimaging results concepts with informal explanations and illustrative examples (e.g. see SPM results). NIDM-Results distinguishes software-agnostic structures, forming the essence of the results, from software-specific structures catering for more specific uses of results by different analysis software. Core and extended structures are respectively presented in Section 2.1 and Section 2.2.

Domain covered by NIDM-Results

NIDM-Results is concerned with the modelling of model fitting and inference in the context of massively univariate analyses. A typical example is the analysis of functional MRI data but studies involving other modalities (such as PET) and sequences (e.g. anatomical MRI through VBM) can also be modelled. The domain covered by NIDM-Results data model is represented in .

Domain overview

NIDM-Results Core Structures

The concepts found in the core of NIDM-Results are introduced in the rest of this section.

Overview

The core NIDM-Results structures are presented in . The color coding corresponds to the prov:type (blue: prov:entity, red: prov:activity). NIDM-Results is divided in two elements: "Model fitting" and "Inference", the structures and relations belonging to each element are presented in details in and .
Core structures overview

Re-use of PROV-DM Structures

PROV relations used in NIDM-Results are summarized in Table 2.

Table 2: PROV relations in use in NIDM-Results
NIDM-Results ConceptsTypes or Relation (PROV concepts)NameOverview
Generation PROV Relations
 WasGeneratedBy PROV-DM wasGeneratedBy
UsageUsedPROV-DM used
DerivationWasDerivedFromPROV-DM wasDerivedFrom

NIDM-Results: Types and relations

General

Table 3: Mapping of NIDM-Results General Core Concepts to types and relations and PROV core concepts
NIDM-Results Concepts Types or Relation (PROV concepts) Name
nidm:Map NIDM-Results Types
(PROV Entity)		 nidm:Map

nidm:Map

nidm:Map: Ordered set of values corresponding to the discrete sampling of some process (e.g. brain MRI data measured on a regular 3D lattice; or brain cortical surface data measured irregularly over the cortex) nidm:Map is a prov:Entity.

A nidm:Map has attributes:

Parameters estimation

Table 4: Mapping of NIDM-Results Parameters estimation Core Concepts to types and relations and PROV core concepts
NIDM-Results Concepts Types or Relation (PROV concepts) Name
nidm:ModelParametersEstimation NIDM-Results Types
(PROV Activity)		 nidm:ModelParametersEstimation
nidm:CustomMaskMap NIDM-Results Types
(PROV Entity)		 nidm:CustomMaskMap
nidm:Data nidm:Data
nidm:DesignMatrix nidm:DesignMatrix
nidm:GrandMeanMap nidm:GrandMeanMap
nidm:MaskMap nidm:MaskMap
nidm:NoiseModel nidm:NoiseModel
nidm:ParameterEstimateMap nidm:ParameterEstimateMap
nidm:ResidualMeanSquaresMap nidm:ResidualMeanSquaresMap

nidm:ModelParametersEstimation

nidm:ModelParametersEstimation: The process of estimating the parameters of a general linear model from the available data nidm:ModelParametersEstimation is a prov:Activity that uses nidm:CustomMaskMap, nidm:Data, nidm:DesignMatrix, nidm:NoiseModel entities. This activity generates nidm:MaskMap, nidm:ParameterEstimateMap, nidm:ResidualMeanSquaresMap, spm:ReselsPerVoxelMap entities. 

activity(niiri:model_pe_id,
      [prov:type = 'nidm:ModelParametersEstimation',
      prov:label = "Model parameters estimation",
      nidm:withEstimationMethod = 'nidm:OrdinaryLeastSquares'
      ])

nidm:EstimationMethod

nidm:EstimationMethod: nidm:EstimationMethod is a prov:Entity.

A nidm:EstimationMethod has attributes:

nidm:CustomMaskMap

nidm:CustomMaskMap: Mask defined by the user to restrain the space in which model fitting is performed nidm:CustomMaskMap is a prov:Entity used by nidm:ModelParametersEstimation.

A nidm:CustomMaskMap has attributes: 
entity(niiri:custom_mask_id_1,
      [prov:type = 'nidm:CustomMaskMap',
      prov:location = "file:///path/to/CustomMask.nii.gz" %% xsd:anyURI,
      nidm:filename = "CustomMask.nii.gz" %% xsd:string,
      dct:format = "image/nifti",
      prov:label = "Custom mask" %% xsd:string,
      nidm:atCoordinateSpace = 'niiri:coordinate_space_id_1',
      crypto:sha512 = "e43b6e01b0463fe7d40782137867a..." %% xsd:string])

nidm:Data

nidm:Data: "A collection or single item of factual information, derived from measurement or research, from which conclusions may be drawn." nidm:Data is a prov:Entity used by nidm:ModelParametersEstimation.

A nidm:Data has attributes:
  • rdfs:label: (OPTIONAL) Human readable description of the nidm:Data.
  • crypto:sha512: (OPTIONAL) (range xsd:string).
  • dct:format: (OPTIONAL) (range xsd:string).
  • nidm:grandMeanScaling: (OPTIONAL) Binary flag defining whether the data was scaled. Specifically, "grand mean scaling" refers to multipliciation of every voxel in every scan by a common value. Grand mean scaling is essential for first-level fMRI, to transform the arbitrary MRI units, but is generally not used with second level analyses (range xsd:boolean).
  • nidm:targetIntensity: (OPTIONAL) Value to which the grand mean of the Data was scaled (applies only if grandMeanScaling is true) (range xsd:float).
entity(niiri:data_id,
      [prov:type = 'nidm:Data',
      prov:type = 'prov:Collection',
      prov:label = "Data" %% xsd:string,
      nidm:grandMeanScaling = "true" %%xsd:boolean,
      nidm:targetIntensity = "100" %% xsd:float])

nidm:DesignMatrix

nidm:DesignMatrix: A matrix of values defining the explanatory variables used in a regression model. Each column corresponds to one explanatory variable, each row corresponds to one observation nidm:DesignMatrix is a prov:Entity used by nidm:ContrastEstimation, nidm:ModelParametersEstimation.

A nidm:DesignMatrix has attributes: 
entity(niiri:design_matrix_id,
      [prov:type = 'nidm:DesignMatrix',
      prov:location = "file:///path/to/DesignMatrix.csv" %% xsd:anyURI,
      dct:format = "text/csv",
      nidm:filename = "DesignMatrix.csv",
      nidm:visualisation = 'niiri:design_matrix_png_id',
      prov:label = "Design Matrix" %% xsd:string])

nidm:Image

nidm:Image: FIXME nidm:Image is a prov:Entity.

A nidm:Image has attributes: 
entity(niiri:design_matrix_png_id,
      [prov:type = 'nidm:Image',
      prov:location = "file:///path/to/DesignMatrix.png" %% xsd:anyURI,
      nidm:filename = "DesignMatrix.png",
      dct:format = "image/png"])
entity(niiri:maximum_intensity_projection_id,
      [prov:type = 'nidm:Image',
      prov:location = "file:///path/to/MaximumIntensityProjection.png" %% xsd:anyURI,
      nidm:filename = "MaximumIntensityProjection.png" %% xsd:string,
      dct:format = "image/png"])

nidm:GrandMeanMap

nidm:GrandMeanMap: FIXME nidm:GrandMeanMap is a prov:Entity.

A nidm:GrandMeanMap has attributes: 
entity(niiri:grand_mean_map_id,
      [prov:type = 'nidm:GrandMeanMap',
      prov:location = "file:///path/to/GrandMean.nii.gz" %% xsd:anyURI,
      nidm:filename = "GrandMean.nii.gz" %% xsd:string,
      dct:format = "image/nifti",
      prov:label = "Grand Mean Map" %% xsd:string,
      nidm:maskedMedian = "115" %% xsd:float,
      nidm:atCoordinateSpace = 'niiri:coordinate_space_id_1',
      crypto:sha512 = "e43b6e01b0463fe7d40782137867a..." %% xsd:string])

nidm:MaskMap

nidm:MaskMap: Map or surface on which the associated results are displayed. nidm:MaskMap is a prov:Entity used by nidm:ContrastEstimation and generated by nidm:ModelParametersEstimation.

A nidm:MaskMap has attributes: 
entity(niiri:mask_id_2,
      [prov:type = 'nidm:MaskMap',
      prov:location = "file:///path/to/Mask.nii.gz" %% xsd:anyURI,
      nidm:filename = "Mask.nii.gz" %% xsd:string,
      dct:format = "image/nifti",
      prov:label = "Mask" %% xsd:string,
      nidm:atCoordinateSpace = 'niiri:coordinate_space_id_1',
      crypto:sha512 = "e43b6e01b0463fe7d40782137867a..." %% xsd:string])

nidm:NoiseModel

nidm:NoiseModel: FIXME nidm:NoiseModel is a prov:Entity used by nidm:ModelParametersEstimation. 

entity(niiri:noise_model_id,
      [prov:type = 'nidm:NoiseModel',
      nidm:hasNoiseDistribution = 'nidm:GaussianDistribution',
      nidm:noiseVarianceHomogeneous = "true" %%xsd:boolean,
      nidm:varianceSpatialModel = 'nidm:SpatiallyLocal',
      nidm:hasNoiseDependence = 'nidm:IndependentNoise',
      nidm:dependenceSpatialModel = 'nidm:SpatiallyLocal'])

nidm:SpatialModel

nidm:SpatialModel: FIXME nidm:SpatialModel is a prov:Entity.

A nidm:SpatialModel has attributes:

nidm:ArbitrarilyCorrelatedNoise

nidm:ArbitrarilyCorrelatedNoise: FIXME nidm:ArbitrarilyCorrelatedNoise is a prov:Entity.

A nidm:ArbitrarilyCorrelatedNoise has attributes:

nidm:CompoundSymmetricNoise

nidm:CompoundSymmetricNoise: FIXME nidm:CompoundSymmetricNoise is a prov:Entity.

A nidm:CompoundSymmetricNoise has attributes:

nidm:ExchangeableNoise

nidm:ExchangeableNoise: FIXME nidm:ExchangeableNoise is a prov:Entity.

A nidm:ExchangeableNoise has attributes:

nidm:IndependentNoise

nidm:IndependentNoise: FIXME nidm:IndependentNoise is a prov:Entity.

A nidm:IndependentNoise has attributes:

nidm:NoiseDependence

nidm:NoiseDependence: FIXME nidm:NoiseDependence is a prov:Entity.

A nidm:NoiseDependence has attributes:

nidm:BinomialDistribution

nidm:BinomialDistribution: nidm:BinomialDistribution is a prov:Entity.

A nidm:BinomialDistribution has attributes:

nidm:GaussianDistribution

nidm:GaussianDistribution: nidm:GaussianDistribution is a prov:Entity.

A nidm:GaussianDistribution has attributes:

nidm:NoiseDistribution

nidm:NoiseDistribution: nidm:NoiseDistribution is a prov:Entity.

A nidm:NoiseDistribution has attributes:

nidm:ParameterEstimateMap

nidm:ParameterEstimateMap: A map whose value at each location is the estimate of a model parameter nidm:ParameterEstimateMap is a prov:Entity used by nidm:ContrastEstimation and generated by nidm:ModelParametersEstimation.

A nidm:ParameterEstimateMap has attributes: 
entity(niiri:parameter_estimate_map_id_1,
      [prov:type = 'nidm:ParameterEstimateMap',
      prov:location = "file:///path/to/ParameterEstimate_0001.nii.gz" %% xsd:anyURI,
      prov:label = "Beta Map 1" %% xsd:string,
      nidm:filename = "ParameterEstimate_0001.nii.gz" %% xsd:string,
      dct:format = "image/nifti",
      nidm:atCoordinateSpace = 'niiri:coordinate_space_id_1',
      crypto:sha512 = "e43b6e01b0463fe7d40782137867a..." %% xsd:string])

nidm:ResidualMeanSquaresMap

nidm:ResidualMeanSquaresMap: A map whose value at each location is the residual of the mean squares fit to the data. nidm:ResidualMeanSquaresMap is a prov:Entity used by nidm:ContrastEstimation and generated by nidm:ModelParametersEstimation.

A nidm:ResidualMeanSquaresMap has attributes: 
entity(niiri:residual_mean_squares_map_id,
      [prov:type = 'nidm:ResidualMeanSquaresMap',
      prov:location = "file:///path/to/ResidualMeanSquares.nii.gz" %% xsd:anyURI,
      nidm:filename = "ResidualMeanSquares.nii.gz" %% xsd:string,
      dct:format = "image/nifti",
      prov:label = "Residual Mean Squares Map" %% xsd:string,
      nidm:atCoordinateSpace = 'niiri:coordinate_space_id_1',
      crypto:sha512 = "e43b6e01b0463fe7d40782137867a..." %% xsd:string])

Contrast estimation

Table 5: Mapping of NIDM-Results Contrast estimation Core Concepts to types and relations and PROV core concepts
NIDM-Results Concepts Types or Relation (PROV concepts) Name
nidm:ContrastEstimation NIDM-Results Types
(PROV Activity)		 nidm:ContrastEstimation
nidm:ContrastMap NIDM-Results Types
(PROV Entity)		 nidm:ContrastMap
nidm:ContrastStandardErrorMap nidm:ContrastStandardErrorMap
nidm:ContrastWeights nidm:ContrastWeights
nidm:StatisticMap nidm:StatisticMap

nidm:ContrastEstimation

nidm:ContrastEstimation: The process of estimating a contrast from the estimated parameters of statistical model nidm:ContrastEstimation is a prov:Activity that uses nidm:ContrastWeights, nidm:DesignMatrix, nidm:MaskMap, nidm:ParameterEstimateMap, nidm:ResidualMeanSquaresMap entities. This activity generates nidm:ContrastMap, nidm:ContrastStandardErrorMap, nidm:StatisticMap entities.

A nidm:ContrastEstimation has attributes: 
activity(niiri:contrast_estimation_id,
      [prov:type = 'nidm:ContrastEstimation',
      prov:label = "Contrast estimation"])

nidm:ContrastMap

nidm:ContrastMap: A map whose value at each location is statistical contrast estimate nidm:ContrastMap is a prov:Entity used by nidm:Inference and generated by nidm:ContrastEstimation.

A nidm:ContrastMap has attributes: 
entity(niiri:contrast_map_id,
      [prov:type = 'nidm:ContrastMap',
      prov:location = "file:///path/to/Contrast.nii.gz" %% xsd:anyURI,
      dct:format = "image/nifti",
      nidm:filename = "Contrast.nii.gz" %% xsd:string,
      prov:label = "Contrast Map: listening > rest" %% xsd:string,
      nidm:contrastName = "listening > rest" %% xsd:string,
      nidm:atCoordinateSpace = 'niiri:coordinate_space_id_1',
      crypto:sha512 = "e43b6e01b0463fe7d40782137867a..." %% xsd:string])

nidm:ContrastStandardErrorMap

nidm:ContrastStandardErrorMap: A map whose value at each location is the standard error of a given contrast nidm:ContrastStandardErrorMap is a prov:Entity generated by nidm:ContrastEstimation.

A nidm:ContrastStandardErrorMap has attributes: 
entity(niiri:contrast_standard_error_map_id,
      [prov:type = 'nidm:ContrastStandardErrorMap',
      prov:location = "file:///path/to/ContrastStandardError.nii.gz" %% xsd:anyURI,
      nidm:filename = "ContrastStandardError.nii.gz" %% xsd:string,
      dct:format = "image/nifti",
      prov:label = "Contrast Standard Error Map" %% xsd:string,
      nidm:atCoordinateSpace = 'niiri:coordinate_space_id_1',
      crypto:sha512 = "e43b6e01b0463fe7d40782137867a..." %% xsd:string])

nidm:ContrastWeights

nidm:ContrastWeights: Vector defining the linear combination associated with a particular contrast. nidm:ContrastWeights is a prov:Entity used by nidm:ContrastEstimation.

A nidm:ContrastWeights has attributes: 
entity(niiri:contrast_weights_id,
      [prov:type = 'nidm:ContrastWeights',
      nidm:statisticType = 'nidm:TStatistic',
      nidm:contrastName = "listening > rest" %% xsd:string,
      prov:label = "Contrast: Listening > Rest" %% xsd:string,
      prov:value = "[1, 0, 0]" %% xsd:string])

nidm:FStatistic

nidm:FStatistic: nidm:FStatistic is a prov:Entity.

A nidm:FStatistic has attributes:

nidm:Statistic

nidm:Statistic: nidm:Statistic is a prov:Entity.

A nidm:Statistic has attributes:

nidm:StatisticMap

nidm:StatisticMap: A map whose value at each location is a statistic. nidm:StatisticMap is a prov:Entity used by nidm:Inference and generated by nidm:ContrastEstimation.

A nidm:StatisticMap has attributes: 
entity(niiri:statistic_map_id,
      [prov:type = 'nidm:StatisticMap',
      nidm:statisticType = 'nidm:TStatistic',
      prov:location = "file:///path/to/TStatistic.nii.gz" %% xsd:anyURI,
      nidm:filename = "TStatistic.nii.gz" %% xsd:string,
      dct:format = "image/nifti",
      prov:label = "Statistic Map: listening > rest" %% xsd:string,
      nidm:contrastName = "listening > rest" %% xsd:string,
      nidm:errorDegreesOfFreedom = "72.9999999990787" %% xsd:float,
      nidm:effectDegreesOfFreedom = "1" %% xsd:float,
      nidm:atCoordinateSpace = 'niiri:coordinate_space_id_1',
      crypto:sha512 = "e43b6e01b0463fe7d40782137867a..." %% xsd:string])

Inference

Table 6: Mapping of NIDM-Results Inference Core Concepts to types and relations and PROV core concepts
NIDM-Results Concepts Types or Relation (PROV concepts) Name
nidm:Inference NIDM-Results Types
(PROV Activity)		 nidm:Inference
nidm:Cluster NIDM-Results Types
(PROV Entity)		 nidm:Cluster
nidm:ClusterLabelsMap nidm:ClusterLabelsMap
nidm:Coordinate nidm:Coordinate
nidm:ExcursionSet nidm:ExcursionSet
nidm:ExtentThreshold nidm:ExtentThreshold
nidm:HeightThreshold nidm:HeightThreshold
nidm:Peak nidm:Peak
nidm:SearchSpaceMap nidm:SearchSpaceMap

nidm:Inference

nidm:Inference: The process of inferring the excursion set from a statistical map nidm:Inference is a prov:Activity that uses nidm:ContrastMap, nidm:StatisticMap, spm:ReselsPerVoxelMap entities.

A nidm:Inference has attributes: 
activity(niiri:inference_id,
      [prov:type = 'nidm:Inference',
      nidm:hasAlternativeHypothesis = 'nidm:OneTailedTest',
      prov:label = "Inference"])

nidm:OneTailedTest

nidm:OneTailedTest: Re-use "one tailed test" from STATO nidm:OneTailedTest is a prov:Entity.

A nidm:OneTailedTest has attributes:

nidm:TwoTailedTest

nidm:TwoTailedTest: Re-use "two tailed test" from STATO nidm:TwoTailedTest is a prov:Entity.

A nidm:TwoTailedTest has attributes:

nidm:Cluster

nidm:Cluster: Statistic defined at the cluster-level in an excusion set. In SPM it is defined as the cluster size. FIXME (now Cluster instead of ClusterStatistic) nidm:Cluster is a prov:Entity derived from nidm:ExcursionSet.

A nidm:Cluster has attributes: 
entity(niiri:cluster_0001,
      [prov:type = 'nidm:Cluster',
      prov:label = "Cluster 0001" %% xsd:string,
      nidm:clusterSizeInVoxels = "530" %% xsd:int,
      nidm:clusterLabelId = 1,
      spm:clusterSizeInResels = "23.1209189500945" %% xsd:float,
      nidm:pValueUncorrected = "9.56276736481136e-52" %% xsd:float,
      nidm:pValueFWER = "0" %% xsd:float,
      nidm:qValueFDR = "7.65021389184909e-51" %% xsd:float])

nidm:ClusterLabelsMap

nidm:ClusterLabelsMap: FIXME nidm:ClusterLabelsMap is a prov:Entity.

A nidm:ClusterLabelsMap has attributes: 
entity(niiri:cluster_label_map_id,
      [prov:type = 'nidm:ClusterLabelsMap',
      prov:location = "file:///path/to/ClusterLabels.nii.gz" %% xsd:anyURI,
      nidm:filename = "ClusterLabels.nii.gz" %% xsd:string,
      dct:format = "image/nifti"])

nidm:Coordinate

nidm:Coordinate: nidm:Coordinate is a prov:Entity.

A nidm:Coordinate has attributes: 
entity(niiri:coordinate_0001,
        [prov:type = 'prov:Location',
        prov:type = 'nidm:Coordinate',
        prov:label = "Coordinate 1" %% xsd:string,
        fsl:coordinate1InVoxels = "45" %% xsd:float,
        fsl:coordinate2InVoxels = "15" %% xsd:float,
        fsl:coordinate3InVoxels = "14" %% xsd:float,
        nidm:coordinate1 = "-48.1" %% xsd:float,
        nidm:coordinate2 = "-73.7" %% xsd:float,
        nidm:coordinate3 = "-9.24" %% xsd:float])
entity(niiri:coordinate_0001,
      [prov:type = 'prov:Location',
      prov:type = 'nidm:Coordinate',
      prov:label = "Coordinate: 0001" %% xsd:string,
      nidm:coordinate1 = "-60" %% xsd:float,
      nidm:coordinate2 = "-28" %% xsd:float,
      nidm:coordinate3 = "13" %% xsd:float])

nidm:ExcursionSet

nidm:ExcursionSet: Set of map elements surviving a thresholding procedure nidm:ExcursionSet is a prov:Entity.

A nidm:ExcursionSet has attributes: 
entity(niiri:excursion_set_id,
      [prov:type = 'nidm:ExcursionSet',
      prov:location = "file:///path/to/ExcursionSet.nii.gz" %% xsd:anyURI,
      dct:format = "image/nifti",
      nidm:filename = "ExcursionSet.nii.gz",
      prov:label = "Excursion Set" %% xsd:string,
      nidm:hasClusterLabelsMap = 'niiri:cluster_label_map_id',
      spm:hasMaximumIntensityProjection = 'niiri:maximum_intensity_projection_id',
      nidm:atCoordinateSpace = 'niiri:coordinate_space_id_1',
      crypto:sha512 = "e43b6e01b0463fe7d40782137867a..." %% xsd:string,
      nidm:numberOfClusters = "8" %% xsd:int,
      nidm:pValue = "8.95949980872501e-14" %% xsd:float])

nidm:ExtentThreshold

nidm:ExtentThreshold: A numerical value that establishes a bound on a range of cluster-sizes. / [from extentThresh:] Minimum cluster size used when thresholding a statistic image 5voxels nidm:ExtentThreshold is a prov:Entity.

A nidm:ExtentThreshold has attributes: 
entity(niiri:extent_threshold_id,
      [prov:type = 'nidm:ExtentThreshold',
      prov:label = "Extent Threshold: k>=0" %% xsd:string,
      nidm:clusterSizeInVoxels = "0" %% xsd:int,
      spm:clusterSizeInResels = "0" %% xsd:float,
      nidm:pValueUncorrected = "1" %% xsd:float,
      nidm:pValueFWER = "1" %% xsd:float])

nidm:HeightThreshold

nidm:HeightThreshold: A numerical value that establishes a bound on a range of voxelwise or vertex-wise defined statistic. nidm:HeightThreshold is a prov:Entity.

A nidm:HeightThreshold has attributes: 
entity(niiri:height_threshold_id,
      [prov:type = 'nidm:HeightThreshold',
      prov:label = "Height Threshold: p<0.05 (FWE)" %% xsd:string,
      nidm:userSpecifiedThresholdType = "p-value FWE" %% xsd:string,
      prov:value = "5.23529984739211" %% xsd:float,
      nidm:pValueUncorrected = "7.62276079258051e-07" %% xsd:float,
      nidm:pValueFWER = "0.05" %% xsd:float])

nidm:Peak

nidm:Peak: Statistic defined at the peak-level in an excursion set. FIXME (now Peak instead of PeakStatistic) nidm:Peak is a prov:Entity derived from nidm:Cluster.

A nidm:Peak has attributes:
  • rdfs:label: (OPTIONAL) Human readable description of the nidm:Peak.
  • crypto:sha512: (OPTIONAL) (range xsd:string).
  • dct:format: (OPTIONAL) (range xsd:string).
  • nidm:equivalentZStatistic: (OPTIONAL) Statistic value transformed into Z units; the output of a process which takes a non-normal statistic and transforms it to an equivalent z score (range xsd:float).
  • nidm:pValueFWER: (OPTIONAL) "A quantitative confidence value resulting from a multiple testing error correction method which adjusts the p-value used as input to control for Type I error in the context of multiple pairwise tests" (range xsd:float).
  • nidm:pValueUncorrected: (OPTIONAL) A p-value reported without correction for multiple testing. (range xsd:float).
  • nidm:qValueFDR: (OPTIONAL) P-value adjusted for the multiple testing, controlling for the False Discovery Rate (range xsd:float).
entity(niiri:peak_0001,
      [prov:type = 'nidm:Peak',
      prov:label = "Peak 0001" %% xsd:string,
      prov:location = 'niiri:coordinate_0001',
      prov:value = "13.9346199035645" %% xsd:float,
      nidm:equivalentZStatistic = "INF" %% xsd:float,
      nidm:pValueUncorrected = "4.44089209850063e-16" %% xsd:float,
      nidm:pValueFWER = "0" %% xsd:float,
      nidm:qValueFDR = "6.3705194444993e-11" %% xsd:float])

nidm:SearchSpaceMap

nidm:SearchSpaceMap: Mask in which the inference was performed / Properties of the underlying statistical process nidm:SearchSpaceMap is a prov:Entity.

A nidm:SearchSpaceMap has attributes: 
entity(niiri:search_space_id,
        [prov:type = 'nidm:SearchSpaceMap',
        prov:location = "file:///path/to/SearchSpace.nii.gz" %% xsd:anyURI,
        dct:format = "image/nifti",
        crypto:sha512 = "400a2f07d99ed9be06577e6ecc89222cf4b688c654bc89067da558e88b73b97dd1b25e6c98f2a735fa0a1409598cff7e6025bda55abb6b9f5ef65d8d307eeba8",
        nidm:filename = "mask.nii.gz" %% xsd:string,
        nidm:filename = "SearchSpace.nii.gz" %% xsd:string,
        nidm:atCoordinateSpace = 'niiri:coordinate_space_id_2',
        prov:label = "Search Space Map" %% xsd:string,
        fsl:searchVolumeInVoxels = "45359" %% xsd:int,
        fsl:reselSizeInVoxels = "12.2251" %%xsd:float])
entity(niiri:search_space_id,
      [prov:type = 'nidm:SearchSpaceMap',
      prov:location = "file:///path/to/SearchSpace.nii.gz" %% xsd:anyURI,
      nidm:filename = "SearchSpace.nii.gz" %% xsd:string,
      dct:format = "image/nifti",
      prov:label = "Search Space Map" %% xsd:string,
      nidm:atCoordinateSpace = 'niiri:coordinate_space_id_2',
      spm:expectedNumberOfVoxelsPerCluster = "0.553331387916112" %% xsd:float,
      spm:expectedNumberOfClusters = "0.0889172687960151" %% xsd:float,
      spm:heightCriticalThresholdFWE05 = "5.23529984739211" %% xsd:float,
      spm:heightCriticalThresholdFDR05 = "6.22537899017334" %% xsd:float,
      spm:smallestSignifClusterSizeInVoxelsFWE05 = "1" %% xsd:int,
      spm:smallestSignifClusterSizeInVoxelsFDR05 = "3" %% xsd:int,
      spm:searchVolumeInVoxels = "65593" %% xsd:int,
      spm:searchVolumeInUnits = "1771011" %% xsd:float,
      spm:reselSize = "22.9229643140043" %% xsd:float,
      spm:searchVolumeInResels = "2552.68032521656" %% xsd:float,
      spm:searchVolumeReselsGeometry = "[3, 72.3216126440484, 850.716735116472, 2552.68032521656]" %% xsd:string,
      spm:noiseFWHMInVoxels = "[2.95881189165801, 2.96628446669584, 2.61180425626264]" %% xsd:string,
      spm:noiseFWHMInUnits = "[8.87643567497404, 8.89885340008753, 7.83541276878791]" %% xsd:string,
      nidm:randomFieldStationarity = "false" %%xsd:boolean,
      crypto:sha512 = "e43b6e01b0463fe7d40782137867a..." %% xsd:string])
entity(niiri:stat_image_properties_id,
      [prov:type = 'nidm:SearchSpaceMap',
      prov:label = "Statistical image properties",
      spm:expectedNumberOfVoxelsPerCluster = "0.553331387916112" %% xsd:float,
      spm:expectedNumberOfClusters = "0.0889172687960151" %% xsd:float,
      spm:heightCriticalThresholdFWE05 = "5.23529984739211" %% xsd:float,
      spm:heightCriticalThresholdFDR05 = "6.22537899017334" %% xsd:float,
      spm:smallestSignifClusterSizeInVoxelsFWE05 = "1" %% xsd:float,
      spm:smallestSignifClusterSizeInVoxelsFDR05 = "3" %% xsd:float])

SPM-specific Concepts

Table 7: Mapping of NIDM-Results SPM-specific Concepts Core Concepts to types and relations and PROV core concepts
NIDM-Results Concepts Types or Relation (PROV concepts) Name
nidm:SPM NIDM-Results Types
(PROV Agent)		 nidm:SPM
spm:ReselsPerVoxelMap NIDM-Results Types
(PROV Entity)		 spm:ReselsPerVoxelMap

nidm:SPM

nidm:SPM: Statistical Parametric Mapping software package for the analysis of neuroimaging data from the FIL Methods Group nidm:SPM is a prov:Agent.

A nidm:SPM has attributes: 
agent(niiri:software_id,
      [prov:type = 'nidm:SPM',
      prov:type = 'prov:SoftwareAgent',
      prov:label = "SPM" %% xsd:string,
      nidm:softwareVersion = "SPM12b" %% xsd:string,
      spm:softwareRevision = "5853" %% xsd:string])

spm:ReselsPerVoxelMap

spm:ReselsPerVoxelMap: A map whose value at each location is the number of resels per voxel. spm:ReselsPerVoxelMap is a prov:Entity used by nidm:Inference and generated by nidm:ModelParametersEstimation.

A spm:ReselsPerVoxelMap has attributes: 
entity(niiri:resels_per_voxel_map_id,
      [prov:type = 'spm:ReselsPerVoxelMap',
      prov:location = "file:///path/to/ReselsPerVoxel.nii.gz" %% xsd:anyURI,
      nidm:filename = "ReselsPerVoxel.nii.gz" %% xsd:string,
      dct:format = "image/nifti",
      prov:label = "Resels per Voxel Map" %% xsd:string,
      nidm:atCoordinateSpace = 'niiri:coordinate_space_id_1',
      crypto:sha512 = "e43b6e01b0463fe7d40782137867a..." %% xsd:string])

FSL-specific Concepts

Table 8: Mapping of NIDM-Results FSL-specific Concepts Core Concepts to types and relations and PROV core concepts
NIDM-Results Concepts Types or Relation (PROV concepts) Name
nidm:FSL NIDM-Results Types
(PROV Agent)		 nidm:FSL
fsl:CenterOfGravity NIDM-Results Types
(PROV Entity)		 fsl:CenterOfGravity

nidm:FSL

nidm:FSL: FMRIB Software Library software package for the analysis of neuroimaging data from the FMRIB nidm:FSL is a prov:Agent.

A nidm:FSL has attributes:

fsl:CenterOfGravity

fsl:CenterOfGravity: Centre Of Gravity for the cluster, equivalent to the concept of Centre Of Gravity for a object with distributed mass, where intensity substitutes for mass in this case (definition from http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/Cluster) fsl:CenterOfGravity is a prov:Entity derived from nidm:Cluster.

A fsl:CenterOfGravity has attributes: 
entity(niiri:center_of_gravity_1,
        [prov:type = 'fsl:CenterOfGravity',
        prov:label = "Center of gravity 1",
        prov:location = 'niiri:COG_coordinate_0001'])