Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Apr;66(4):507-519.
doi: 10.1007/s00234-024-03304-3. Epub 2024 Feb 21.

Removing outliers from the normative database improves regional atrophy detection in single-subject voxel-based morphometry

Vivian Schultz  1 , Dennis M Hedderich  2 , Benita Schmitz-Koep  2 , David Schinz  2   3 , Claus Zimmer  2 , Igor Yakushev  4 , Ivayla Apostolova  5 , Cansu Özden  5 , Roland Opfer  6 , Ralph Buchert  5
Affiliations

Removing outliers from the normative database improves regional atrophy detection in single-subject voxel-based morphometry

Vivian Schultz et al. Neuroradiology. 2024 Apr.

Abstract

Purpose: Single-subject voxel-based morphometry (VBM) compares an individual T1-weighted MRI to a sample of normal MRI in a normative database (NDB) to detect regional atrophy. Outliers in the NDB might result in reduced sensitivity of VBM. The primary aim of the current study was to propose a method for outlier removal ("NDB cleaning") and to test its impact on the performance of VBM for detection of Alzheimer's disease (AD) and frontotemporal lobar degeneration (FTLD).

Methods: T1-weighted MRI of 81 patients with biomarker-confirmed AD (n = 51) or FTLD (n = 30) and 37 healthy subjects with simultaneous FDG-PET/MRI were included as test dataset. Two different NDBs were used: a scanner-specific NDB (37 healthy controls from the test dataset) and a non-scanner-specific NDB comprising 164 normal T1-weighted MRI from 164 different MRI scanners. Three different quality metrics based on leave-one-out testing of the scans in the NDB were implemented. A scan was removed if it was an outlier with respect to one or more quality metrics. VBM maps generated with and without NDB cleaning were assessed visually for the presence of AD or FTLD.

Results: Specificity of visual interpretation of the VBM maps for detection of AD or FTLD was 100% in all settings. Sensitivity was increased by NDB cleaning with both NDBs. The effect was statistically significant for the multiple-scanner NDB (from 0.47 [95%-CI 0.36-0.58] to 0.61 [0.49-0.71]).

Conclusion: NDB cleaning has the potential to improve the sensitivity of VBM for the detection of AD or FTLD without increasing the risk of false positive findings.

Keywords: Brain; Magnetic resonance imaging; Neurodegeneration; Normative database; Voxel-based-morphometry.

PubMed Disclaimer

Conflict of interest statement

We declare that we have no conflict of interest.

Figures

Fig. 1
Fig. 1
Standard display for visual interpretation of VBM maps. The example shows the VBM map of a 66-year-old patient with posterior cortical atrophy obtained with the full single-scanner normative database (SSD)
Fig. 2
Fig. 2
Voxel-wise mean of the GM density in the single-scanner normative database (SSD) (top) and in the multiple-scanner normative database (MSD) (bottom) before (left) and after (right) removal of outliers. Mean value ± standard deviation (range) is given for each setting
Fig. 3
Fig. 3
Voxel-wise standard deviation of the GM density in the single-scanner normative database (SSD) (top) and in the multiple-scanner normative database (MSD) (bottom) before (left) and after (right) removal of outliers. Mean value ± standard deviation (range) is given for each setting. The maximum of the color table was set to 0.15
Fig. 4
Fig. 4
Mean value and standard error (SE) of the total volume of atrophy with the scanner-specific normative database (SSD) and with the multiple-scanner normative database (MSD) without and with removal of outliers (“cleaning”) in healthy controls (HC), patients with Alzheimer’s disease (AD), and patients with frontotemporal lobar degeneration (FTLD)
Fig. 5
Fig. 5
VBM maps of a patient with reference standard diagnosis of mild cognitive impairment (MCI) due to Alzheimer’s disease (AD) with (top-to-bottom) the scanner-specific normative database (SSD) before and after removal of outliers (“cleaning”) and with the multiple-scanner normative database (MSD) before and after removal of outliers. Removal of outliers led to a better delineation of hippocampal atrophy with the MSD whereas multiple (unspecific) atrophy clusters were detected with the MSD before removal of outliers as well as with the SSD independent of removal of outliers. The between-readers consensus of the visual interpretation was false negative (“no neurodegenerative disease”) with the SSD before and after removal of outliers and the MSD before removal of outliers and true positive (“neurodegenerative disease (AD)”) after removal of outliers with the MSD
Fig. 6
Fig. 6
Histogram of the change in the total volume of atrophy by removal of outliers from the multiple-scanner normative database (MSD). The corresponding change in the consensus binary visual interpretation (presence versus absence of neurodegeneration) of the VBM maps is indicated by different colors
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

  • Practical brain MRI guidelines for anti-Aβ antibody treatment in early symptomatic Alzheimer's disease.
    Kakeda S, Miki Y, Kudo K, Mori H, Tokumaru AM, Abe O, Aoki S; Working Group for Brain MRI Guidelines for Anti-Aβ Antibody Treatment from The Japan Radiological Society, The Japanese Society for Neuroradiology, and The Japanese Society for Magnetic Resonance in Medicine. Kakeda S, et al. Jpn J Radiol. 2025 Aug;43(8):1231-1238. doi: 10.1007/s11604-025-01773-x. Epub 2025 Apr 23. Jpn J Radiol. 2025. PMID: 40266549 Free PMC article.

References

    1. Ashburner J, Friston KJ. Voxel-based morphometry—the methods. Neuroimage. 2000;11:805–821. doi: 10.1006/nimg.2000.0582. - DOI - PubMed
    1. Huang H, Zheng S, Yang Z et al (2022) Voxel-based morphometry and a deep learning model for the diagnosis of early Alzheimer’s disease based on cerebral gray matter changes. Cereb Cortex 10.1093/cercor/bhac099 - PMC - PubMed
    1. Hedderich DM, Dieckmeyer M, Andrisan T, et al. Normative brain volume reports may improve differential diagnosis of dementing neurodegenerative diseases in clinical practice. Eur Radiol. 2020;30:2821–2829. doi: 10.1007/s00330-019-06602-0. - DOI - PubMed
    1. Bruun M, Frederiksen KS, Rhodius-Meester HFM, et al. Impact of a clinical decision support tool on prediction of progression in early-stage dementia: a prospective validation study. Alzheimers Res Ther. 2019;11:25. doi: 10.1186/s13195-019-0482-3. - DOI - PMC - PubMed
    1. Potvin O, Dieumegarde L, Duchesne S, AsDN I. Normative morphometric data for cerebral cortical areas over the lifetime of the adult human brain. Neuroimage. 2017;156:315–339. doi: 10.1016/j.neuroimage.2017.05.019. - DOI - PubMed

MeSH terms