Harmonizing machine learning imaging biomarkers – CDESH and the case for harmonization in both MR image and low-dimensional (scalar) output space

Jeffrey L. Gunter, Petrice M. Cogswell, Matthew L. Senjem, Kejal Kantarci, David S Knopman, Ronald C. Petersen, Neill R. Graff-Radford, Jonathan Graff-Radford, Clifford R Jr. Jack

Research output: Contribution to journalComment/debatepeer-review

Abstract

Background: In 2018 we presented the Computational DESH (CDESH) metric (DOI:10.1016/j.nicl.2018.11.015), a machine learning measure of abnormal CSF space patterns commonly observed on MRI in Disproportionately Enlarged Subarachnoid-space Hydrocephalus (DESH-type) NPH. We have shown that CDESH positive individuals often have discordant CSF and PET amyloid measures (DOI:10.1016/j.jalz.2019.06.4921). Positive CDESH scores are predictive of cognitive decline (DOI:10.1016/j.neuroimage.2021.117899, DOI: 10.1212/WNL.0000000000008616). The score is calculated from regional CSF volumes in a sulcal atlas and within the ventricles using a support vector machine (SVM). The Mayo Clinic Study of Aging (MCSA) changed scanner platforms (GE to Siemens) and T1-weighted image resolution (1.2mm x 1.05mm x 1.05mm to 0.8mm isotropic) in 2017. We refer to these as “isotropic” or “anisotropic” scans. Our objective was to assess the dependence of a machine learning method (specifically CDESH) to image input variation and, if necessary, determine what needed corrections. Method: All data considered is from the MCSA or Mayo Clinic ADRC. The SVM was trained on anisotropic images with laboriously labelled ground truth. We resampled the isotropic data from a crossover study (N=112, typically scanned same-day with both platforms/resolutions) to match the anisotropic resolution. Two approaches to correction were evaluated (Figure 1). In Figure 2 we compare of the score with and without correction cross-sectionally in 998 MCSA participants. In Figure 3 we present longitudinal series from 23 participants with multiple scans using the isotropic resolution protocol and at least one prior anisotropic image. Results: Comparisons of data from the crossover study are presented in Figure 1. Matching voxel resolution to the data used to train the SVM is critical and a subsequent mild linear score adjustment may be warranted. Cross-sectional comparisons (Figure 2) confirm improvement matching voxel resolution with a possible mild linear score adjustment. After matching voxel resolution CDESH score trajectories (Figure 3) are markedly more stable across this scanner/protocol change. Conclusion: The SVM-based CDESH score is sensitive to voxel resolution. Corrections are needed in image processing and possibly in the scalar output score to achieve stability across protocol and scanner changes. The resulting correction supports coherent long timeseries data and longitudinal extension of previous work.

Original languageEnglish (US)
Article numbere067965
JournalAlzheimer's and Dementia
Volume18
Issue numberS1
DOIs
StatePublished - Dec 2022

ASJC Scopus subject areas

  • Epidemiology
  • Health Policy
  • Developmental Neuroscience
  • Clinical Neurology
  • Geriatrics and Gerontology
  • Cellular and Molecular Neuroscience
  • Psychiatry and Mental health

Fingerprint

Dive into the research topics of 'Harmonizing machine learning imaging biomarkers – CDESH and the case for harmonization in both MR image and low-dimensional (scalar) output space'. Together they form a unique fingerprint.

Cite this