A review of the general approach is Ardekani 2019. This FSL-based pipeline uses an existing Freesurfer hippocampal segmentation (e.g. from https://github.com/baxpr/freesurfer720) to identify hippocampal tissue.
-
Initial 12-parameter affine registration of T1 image to MNI atlas. The freesurfer brainmask in subject space and the MNI brainmask in atlas space, both dilated, are used to improve accuracy.
-
Additional nonlinear registration (warp) to the atlas following default FNIRT 'T1_2_MNI152_2mm' schedule file.
-
Transform the Harvard-Oxford subcortical probabilistic atlas to the native T1 space and threshold the hippocampus maps at the specified level (default of 25%) to obtain an atlas hippocampus mask in the native space.
-
Use the 1st percentile of intensity in the Freesurfer segmented hippocampus regions to obtain a threshold for tissue vs CSF.
-
The reported hippocampal parenchymal fraction is the percentage of voxels in the native space atlas ROI that are at or above the tissue intensity threshold.
-
HPF is reported for both the atlas regions obtained by the affine transform, and those obtained by the nonlinear warp transform.
Ardekani BA, Hadid SA, Blessing E, Bachman AH. Sexual Dimorphism and Hemispheric Asymmetry of Hippocampal Volumetric Integrity in Normal Aging and Alzheimer Disease. AJNR Am J Neuroradiol. 2019 Feb;40(2):276-282. doi: 10.3174/ajnr.A5943. Epub 2019 Jan 17. PMID: 30655257; PMCID: PMC7028613. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7028613/
Hippocampal parenchymal fraction https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7028613/
Registration algo: https://www.biorxiv.org/content/10.1101/306811v1.full Software: https://www.nitrc.org/projects/art (KAIBA)
Refs that compare vs other methods:
22 Ardekani 2016 Accuracy 97% vs 89% for FS in classifying early AD Analysis of the MIRIAD Data Shows Sex Differences in Hippocampal Atrophy Progression https://pubmed.ncbi.nlm.nih.gov/26836168/
23 Goff 2018 Successful detection of volume change in FEP vs FS; better test-retest in controls Association of Hippocampal Atrophy With Duration of Untreated Psychosis and Molecular Biomarkers During Initial Antipsychotic Treatment of First-Episode Psychosis https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5875378/
25 Bruno 2016 Better predictor of future decline than volume Hippocampal volume and integrity as predictors of cognitive decline in intact elderly https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4929020/
Reference, data set, metric of comparison, result
Bruno 2016, Bruno 2016, SPM volumetry, p=0.0003 vs p=0.07 Ardekani 2016, MIRIAD, FS5, Cohen D 3.4 vs 3.1, 2.8 vs 1.2 Goff 2018, Goff 2018, FS6 longitudinal, p=0.001 vs p=0.10 Goff 2018, Maclaren 2014, FS6 longitudinal, CoV 0.4% vs 1.0%
Maclaren J, Han Z, Vos SB, Fischbein N, Bammer R. Reliability of brain volume measurements: a test- retest dataset. Scientific data. 2014;1:140037.