Senjem, Matthew L.

Abstract Background Neuroimaging data typically arrives as Digital Imaging and Communications in Medicine (DICOM) files, batched by exam. We use exam or study to denote a single imaging session of a participant, containing up to ∼40 series (the generic term for scans and derived products). Different series have different properties and applications, making their labeling complex. Filenames often change in transit, and Series Description tags (0008, 103E) are arbitrary, operator‐entered, and unreliably coupled to the series type. Furthermore, label preferences vary from site to site, making the application of a common classification system mandatory for multisite studies. Here we describe the system that we have developed and applied to the Alzheimer's Disease Neuroimaging Initiative (ADNI) and SCAN. Method Although Series Description is unreliable, DICOM includes other tags that parameterize MR pulse sequences and suggest their purpose. The mapping from these features to types is difficult to explicitly specify, especially since different manufacturers and protocols use different conventions. Thus we developed a machine learning classifier using 4060 semiautomatically labeled head and spinal series from 1.5 to 7T MR scanners, including both site‐specific and standard protocols designed to study a wide range of neurological conditions, including dementia, hydrocephalus, and multiple sclerosis. An 80%/20% split, stratified by type, was used for the training/test sets. After initial training, a few cases in the training set with high errors due to rarity were augmented and the model retrained. Performance was evaluated on the clean test set. Result The taxonomy is presented in Table 1. The classifier correctly identified 807 out of 812 (99.4%) series, with the error cases coming from outside the ADNI or SCAN protocols (Table 2). The confusion matrix is shown in Figure 1. Conclusion We implemented an automatic series type classifier with labels chosen to be generally useful for the brain MRI community. Some errors were noted, but the test set was deliberately challenging and the errors appear to be ones that could easily be made by a human. The classifier is expected to have good tolerance for as yet unseen head MR protocols but is likely unsatisfactory for body MRI.

Abstract Background The Overlap Index (OI) as previously published (Lee et al., 2022) is a reliable method for detecting tau accumulation using tau‐PET imaging (flortaucipir; AV1451). It identifies voxel‐wise standardized uptake value ratio (SUVr) elevations across serial scans. Although OI showed more sensitivity to tau signal elevation and longitudinal change than standard ROI measures, to our knowledge, the relationship of the tau PET overlap index to neuropathologic findings is still unclear. This study aims to investigate the potential of OI to detect NFTs of the neuropathologic Braak tangle stage, especially the early Braak tangle stage (I–IV). Method This study included 57 participants from the Mayo Clinic Study of Aging or Mayo Alzheimer's Disease Research Center, all of whom underwent autopsy and at least two serial AV1451 tau‐PET scans and 3T T1‐weighted MRI within an average of 16 months. Tau‐PET SUVr images were normalized to cerebellar crus grey matter, and MR images were co‐registered to a single‐subject MRI mean. Tau‐PET scans were resampled into MRI space, and OI was calculated by applying intensity thresholds 1.3. Clusters with fewer than 20 contiguous voxels were excluded. An OI cutoff of 0.5 was used and an SUVr cutoff of 1.29. Postmortem immunohistochemistry (IHC) using phospho‐tau antibodies (AT8, PHF‐1) was performed on the amygdala, temporal cortex (middle and superior), and anterior/posterior hippocampus. Neuropathological Braak staging was determined based on AT8‐stained sections. We conducted visual comparison of tau‐PET overlap signal in these regions with IHC‐confirmed tau pathology across Braak tangle stages. Result OI‐detected voxels corresponded to NFT distribution across Braak stages. In early Braak stages, standard SUVr‐based META‐ROI values remained negative, whereas OI showed positive signals, indicating greater sensitivity to early tau pathology (Figure 1). The spatial distribution of OI‐derived overlap regions aligned with high NFT burden, suggesting that OI captures early‐stage tau pathology more effectively than conventional SUVr methods. Conclusion OI in serial tau‐PET imaging demonstrated better detection of early‐stage tau pathology than standard ROI‐based SUVr measures. The alignment between OI‐detected voxels and neuropathological tau burden supports its potential as an in vivo biomarker for detecting early neurofibrillary tangles.