Citations

Abstract Background Neuropathologic evaluation is essential for diagnosing and treating dementia, but definitive diagnoses is only available after post‐mortem brain autopsy. Positron emission tomography (PET) offers a potential in vivo proxy for predicting neuropathology. This study proposes a semi‐supervised learning (SSL) model using a single FDG‐PET image to predict underlying neuropathology. Method The SSL model leverages a self‐organized graph structure based on a patient similarity network (PSN) created from 1,495 FDG‐PET images across the aging and neurodegenerative spectrum. Each patient is represented by a node, with edges indicating image similarity derived from principal component analysis features and cosine similarity weights. The PSN structure connects 20 nearest neighbors. Neuropathologic labels were available for 204 autopsied patients. The Poisson Learning Method was employed to propagate labels from these annotated nodes to unlabeled nodes, guided by the PSN structure. FDG‐PET images were collected from clinical and research participants at the Mayo Clinic and processed using an MRI‐free pipeline. Neuropathologic labels were assigned based on widely accepted diagnostic criteria. Leave‐one‐out cross‐validation (LOOCV) was used to assess model performance. Results The confusion matrix demonstrated high accuracy in predicting Alzheimer's disease, frontotemporal lobar degeneration with tau inclusions, and Lewy body disease pathology. A Sankey diagram and label distribution analysis revealed strong correlations between clinical and predicted neuropathologic labels. Conclusion This study highlights the potential of a graph‐based SSL pipeline to predict neuropathological changes using FDG‐PET imaging data, despite limited pathology labels. The findings underscore the value of SSL in bridging the gap between abundant neuroimaging data and scarce neuropathologic labels, contributing to improved dementia diagnosis and understanding.

Abstract Background Multi‐site aging and dementia studies are challenged by data heterogeneity caused by scanner (manufacturer and protocol) related variability. Quantitative analysis of white matter hyperintensity (WMH) from pooled samples, a traditional marker of cerebrovascular disease, is sensitive to machine variability. Our goal was to compare Deming (symmetric) linear regression and ComBat, a common method used for harmonizing neuroimaging measurements, for harmonizing WMH measurements on participants scanned on two different scanners (GE and Siemens) and protocols (2D FLAIR on GE and 3D FLAIR on Siemens). Methods Cognitively unimpaired (CU) ( n =53, age (mean (SD))=67 (15) years) and cognitively impaired (CI; MCI/dementia) ( n =34, age (mean (SD))=82 (9) years) participants from the Mayo Clinic Study of Aging and Mayo ADRC were scanned on GE and Siemens scanners within a week of each other (typically done on the same day). WMH as a percentage of total intracranial volume was measured using multispectral segmentation of FLAIR and T1‐weighted images (Gunter et. al. AAIC 2023). We used Deming regression as it includes errors in both input dimensions. ComBat was applied using age, sex, and diagnosis as covariates, with scanner vendor and protocol as the batch variable. We assessed the inter‐scanner similarity using Maximum Mean Discrepancy (MMD) analysis, where a value of zero indicates identical distributions, p >0.05 indicates no significant difference under H 0 . (Gretton et al., JMLR 2012). Results Our results showed that the distribution of WMH measurements, which were different in the CU group, were more similar after Deming than ComBat harmonization (Figure 1). This was confirmed by the lower MMD in the CU group from Deming regression (‐0.01 ( p =0.76) from 0.03 ( p =0.05)) compared to ComBat (0.04 ( p =0.02))(Figure 1). However, ComBat and Deming were unable to create distributions as compatible as original measurements for the CI group. Conclusion Using a unique dataset of the same participants scanned close in time on different scanner and protocols, we found that (1) MMD measurements were useful to test the need/improvement due to harmonization; (2) ComBat was not substantially better at creating compatible distributions between scanner vendors than Deming regression, and both were sometimes inadequate. Future work will require alternative models to linear scale and shift transformations.

Abstract Background Ten years after the original publication, the Centiloid framework is now broadly used to harmonize amyloid‐PET quantification, facilitate data sharing and comparison across cohorts, and even assist visual interpretation in clinical settings. We evaluated the global implementation of Centiloids by comparing their distribution and corresponding positivity thresholds across cohorts. Methods We gathered data from publicly available cohorts and reached out to investigators across the world to collect cross‐sectional Centiloids, demographic and clinical information, and visual read data. Gaussian mixture models (GMM, k=2) were fitted to Centiloid values for each cohort and cutoffs were calculated as mean + 2SD of the lower Gaussian. When visual reads were available, we determined Centiloid cutoffs that maximized correspondence with visual reads (Cohen’s kappa). Data was combined across cohorts using random effects meta‐analyses. Results As of January 2025, we included 37 cohorts ( n = 41,678 participants) with heterogeneous pipelines, radiotracers, and clinical and demographic characteristics (Table‐1). The low Gaussian peaks ranged from ‐9 to 10CL; the meta‐analysis identified a common peak at 1CL. The second peaks were more heterogeneous (range=38‐102CL; meta‐analysis outcome=64CL). Across cohorts, the proportion of cognitively unimpaired versus impaired participants impacted the position of both peaks, with better separation in cohorts enriched in impaired individuals (Figure‐1C). The meta‐analysis indicated a GMM‐based cutoff of 19CL (95%CI: 16‐21CL, Figure‐1B); subgroup analyses showed no evidence of significant effect between single versus multicenter settings (17 versus 20CL, p = 0.30), MRI‐based or PET‐only processing (18 versus 19CL, p = 0.88), and no evidence of difference across radiotracers (Flutemetamol: 16CL; PIB: 17CL, Flutafuranol: 18CL, Florbetaben: 19CL, Florbetapir: 20CL, p = 0.78). In a subset of 29,496 participants with visual reads available, binary visual reads corresponded well to Centiloids (common kappa=0.86, Figure‐2A). The visual read‐based cutoff of 24CL (95%CI: 21‐27CL, Figure‐2B) maximized correspondence between visual read and quantification and was slightly higher than the GMM‐based cutoff. All meta‐analysis models showed high non‐random heterogeneity (I 2 >80%) across studies, suggesting non‐random differences in peaks and cutoffs. Conclusions Meta‐analysis‐based cutoffs align well with thresholds from the existing literature. High heterogeneity among studies underscores the need to investigate contributing factors, raising concerns about applying common cutoffs.

Abstract Background Use of software to “de‐face” research brain images has grown as advances in automated face recognition have made it increasingly possible to re‐identify participants. Imaging studies of Alzheimer’s Disease and related diseases have included more participants from historically underrepresented groups, but de‐facing software have primarily been validated using images from mostly non‐Hispanic White participants, and its performance has not been directly compared across ethnoracial groups. Method From The Health and Aging Brain Study: Health Disparities (HABS‐HD) study, we constructed a sample of 305 T1‐weighted MRI from HABS‐HD baseline visits, matched across three ethnoracial groups (non‐Hispanic White (NHW, n = 104), non‐Hispanic African American (AA, n = 104), Mexican American (MA, n = 97)) according to age‐decade, sex, cognitive status (impaired/unimpaired), and MRI scanner model. We ran mri_reface 0.3.5 on each image, and we ran FreeSurfer 7.4.1 on each unmodified image and on each de‐faced output. We then compared regional gray matter (GM) volume and cortical thickness measurements by subtracting each de‐faced image measurement from the corresponding unmodified image measurement. We compared these differences between de‐faced and unmodified images across the three matched groups using GLM analyses adjusted for age, sex, and cognitive status, and Bonferroni corrected. Result When aggregating across all brain regions, differences in cortical thickness measures due to de‐facing were significantly different between the NHW and MA groups, and between AA and MA, but not between NHW and AA (Figure 1). Differences between these group means were <0.005mm. For GM volume, differences were significant between AA and MA (magnitude < 0.2%), but not between NHW and AA or NHW and MA (Figure 2). Examining each region individually (Figure 3), cortical thickness differences were significant for only 4 regions (all with magnitudes <0.03mm), and GM volume differences were significant for only 6 regions (all with magnitudes <2%). Conclusion Effects of mri_reface on FreeSurfer brain measurements were significantly different across ethnoracial groups. Our future work will optimize mri_reface to eliminate these differences. However, the magnitudes of these differences are sufficiently small that they are unlikely to have a practical effect on analyses of data using current versions.

Abstract Background Dementia with Lewy Bodies (DLB) is often misdiagnosed or conflated with Alzheimer’s Disease (AD) due to overlapping clinical presentations and neuropathological features. The presence of AD‐like features correlates with accelerated cognitive decline and a worse overall prognosis. Compared to AD, studies suggest that amyloid uptake in DLB may spare the occipital lobe. This project aims to investigate whether amyloid PET could be leveraged to differentiate between AD and DLB patients, with the goal of improving diagnostic accuracy and supporting patient stratification and safety in anti‐amyloid trials. Method This study involves collaboration across six centres (University of Geneva, BioFINDER, AMPLE, AMYPAD PNHS, MCSA, GEHC), each contributing amyloid PET imaging data from patients diagnosed with DLB based on clinical or neuropathological assessments. Moreover, each center provided an additional subset of AD and healthy controls, then matched to DLB samples based on age/sex/APOE genotype/education for comparison. A sub analysis will be performed on patients with evidence of both DLB and AD markers. Given the relatively small sample size, key aspect of the methodological approach is to reduce technical sources of variability. The processing workflow includes rigorous quality control, formatting, normalisation to MNI space, harmonisation (differential smoothing), and MRI‐based segmentation. Result The dataset currently includes amyloid PET scans (acquired using [ 11 C]PiB, [ 18 F]Flutemetamol or [ 18 F]Florbetapir) and MRI scans from ∼130 DLB patients, along with ∼150 AD patients and ∼150 controls. In addition, 99 patients positive for the CSF α‐Synuclein Seeding Amplification Assay were included. Patient data were anonymized, and a harmonised database was created, including demographic characteristics, clinical history, and results from neuropsychological assessments. Conclusion The creation of a standardized, multi‐centre amyloid PET dataset for DLB will enable the study of amyloid PET patterns at the global, regional and voxel level, in comparison with uptake in AD patients and controls. This dataset will serve as a valuable resource for investigating the neuropathological underpinnings of DLB, improving diagnostic accuracy, and potentially guiding patient stratification in anti‐amyloid trials. Efforts to integrate [ 18 F]Florbetaben and additional data are ongoing.

Abstract Background The Joint Propagation Model (JPM)‐based CenTauR scale was recently introduced to harmonize tau‐PET quantification across different radiotracers. This study examined how CenTauR harmonization enhances the comparability of tau‐PET quantification across matched cohorts using different tracers. The evaluation focused on three key aspects: (1) comparing tau‐PET positivity rates as defined by CenTauR, (2) evaluating its diagnostic accuracy for symptomatic AD, and (3) analyzing longitudinal changes in tau‐PET rates over time. Method The JPM, developed by the CPAD‐led Tau PET Harmonization Working Group (Leuzy et al., Alzheimers Dement. 2024; Figure 1A), models relationships between anchor point subjects and head‐to‐head tau‐PET SUVR data onto the CenTauR scale, providing conversion equations for multiple tracers. JPM equations were applied to [¹⁸F]flortaucipir SUVR (Meta‐temporal ROI) data from 561 cognitively impaired participants in the ADNI and OASIS‐3 studies. Using ROC analysis, we determined the CenTauR cut‐off for positivity (T+) that maximized the Youden index for discriminating between FDA‐approved positive/negative visual reads. Three separate cohorts, scanned using [¹⁸F]flortaucipir (ADNI), [¹⁸F]MK‐6240 (CPAS), and [¹⁸F]RO‐948 (BioFINDER‐2), were matched 1:1 by age, MMSE, amyloid‐β, and clinical diagnosis. CenTauR‐based metrics were compared across these cohorts. Result A cut‐off of 18 CenTauRs was found to optimally classify [ 18 F]flortaucipir PET visual reads (Figure 1B). The matching procedure identified 1089 participants (363 per cohort). The frequency of T+ using the 18 CenTauRs cut‐off for the Meta‐Temporal ROI was highly comparable across tracers across groups, except for the Aβ‐positive cognitively unimpaired group, where variability was more pronounced due to a smaller sample size (Figure 2A). Similarly, tau‐PET discriminative accuracy for symptomatic AD vs controls remained similar across [ 18 F]flortaucipir and [ 18 F]MK‐6240 (Figure 2B). In a separate sample of 212 matched participants from ADNI ([¹⁸F]flortaucipir) and AIBL ([¹⁸F]MK‐6240) with baseline and 1‐year follow‐up tau‐PET scans, 1‐year change in CenTauRs was comparable (Figure 2C). An exploratory voxelwise transformation, utilizing the Meta‐Temporal conversion equation in a representative case scanned with both [¹⁸F]flortaucipir and [¹⁸F]MK‐6240, demonstrates the potential for voxelwise CenTauR harmonization (Figure 3). Conclusion These analyses suggest that CenTauR harmonization increases the comparability of tau‐PET data acquired with different tracers. Additional validation analyses with larger cohorts including other radiotracers ([¹⁸F]PI‐2620) are underway.

Study Design. Modified Delphi consensus process Objective. To establish a standardized, consensus-driven definition for postoperative spine infection (PSI) for research, diagnosis, and management. Summary of Background Data. Postoperative infection is a devastating and common complication following spine surgery. Rates of PSI in the literature range from 0% to over 20% depending on surgical indication and invasiveness. Despite the implications of PSI, there is no universally accepted diagnostic standard or consensus definition for infection following spine surgery, hindering research and treatment of this challenging clinical condition. Methods. A multispecialty workgroup convened by the Musculoskeletal Infection Society (MSIS) conducted a systematic literature review. Using data from the systematic review, an expert panel completed a modified Delphi process to achieve consensus. The panel included nine fellowship-trained, board-certified physicians with expertise in infectious diseases, orthopaedic and neurologic spine surgery, and musculoskeletal radiology. Iterative rounds focused on criterion selection, importance ranking, categorical grouping, and final constellation development for PSI diagnosis. Consensus was defined as a 2/3 majority, with a target of 100%. Results. The workgroup achieved 100% consensus on a new PSI definition, which incorporates six clinical domains: wound features, microbiology, imaging, inflammatory biomarkers, intraoperative findings, and histology. A single microbiological stand-alone criterion—identification of a phenotypically indistinguishable organism from two or more deep operative site specimens—was established as pathognomonic for postoperative spine infection. Additional primary and secondary supporting criteria were defined, with specific combinations required to classify cases as definite or probable PSI. Final definitions were ratified by the boards of MSIS and the European Bone & Joint Infection Society (EBJIS) Conclusion. This consensus-based definition provides a standardized framework for diagnosing postoperative spine infection, facilitating research and clinical management. The criteria balance sensitivity and specificity across diverse clinical scenarios and represent the first multispecialty, society-endorsed definition for PSI.

Background: Patients with abdominal aortic aneurysms (AAA) often have comorbidities that make them poor candidates for open surgical repair (OSR). Endovascular aortic aneurysm repair (EVAR) circumvents the morbidity associated with OSR of AAA. Candidacy for EVAR is impacted by multiple factors, including the quality and size of vascular access vessels and involvement of visceral arteries, as seen in paravisceral abdominal aortic aneurysms (PVAAA). Additional challenges, such as obesity, can complicate vascular access during EVAR. PVAAA may be better suited for fenestrated EVAR (FEVAR) using a physician-modified endograft (PMEG). Materials and Methods: In this report, we describe the case of a 64-year-old female patient with a complex PVAAA in the setting of significant aortoiliac occlusive disease (AIOD) with intermittent claudication. We describe a successful staged FEVAR technique in which vascular access challenges were first addressed by recanalization of the iliac system with femoral conduit (FC) creation, followed by successful FEVAR using bilateral FC for deployment of the PMEG. Conclusion: Patients with PAAA have unique and complex pathology that are clinically challenging to address. It is not uncommon that these patients also have co-morbid conditions that make them less-than ideal candidates for open repair. Patients may also have other conditions such as AIOD which make EVAR complex. Staged approach with iliac recanalization and femoral conduit creation followed by FEVAR with PMEG is an effective treatment option for high-risk complex patients and can help avoid common complications such as groin infection and delay in FEVAR which may result in a catastrophic event such as interval AAA rupture.

Hearing loss is a highly prevalent disease state affecting almost 1 in 5 individuals globally. Growing research shows that untreated or undertreated hearing loss carries significant long-term sequelae, including increased risk of social isolation, depression, and cognitive decline, among other ramifications. Limited awareness and common misperceptions among the general US adult population and primary care providers contribute to low rates of referral to hearing specialty clinics and underutilization of hearing aids, cochlear implants, and other rehabilitative devices. Comprised of diverse membership, the Hearing Health Collaborative (HHC) was formed in May 2021 to identify and overcome challenges impacting good health care practices and public policy on hearing care in connection with healthy aging. This editorial documents a recent outcome of the HHC derived from a structured A3 process. Specifically, initiative 2, a proposal to develop a simple, clinically anchored staging system primarily for use by frontline providers and patients, is presented. Unlike other common disorders that harbor long-term health sequelae, a corresponding pragmatic clinical staging system—leveraging precedent disease staging convention, based on clinically meaningful outcomes and prognosis for acquired hearing loss—does not exist. The development of this staging system will potentially allow for improved communication across providers and patients, facilitate referral pathways, and ultimately enable patient access to timely and appropriate care.

Background: Adeno-associated viruses (AAVs) are widely used vectors for in vivo gene therapy, but their standard storage at −80 °C limits deployment in regions lacking ultracold infrastructure. Strategies enabling stable AAV storage at higher temperatures are needed to support global distribution. Methods: Nine AAV serotypes were lyophilized in simple sucrose-based buffers. Post-lyophilization vector integrity was assessed by measuring in vitro transduction efficiency using a luciferase reporter in cell-based assays. Stability of selected serotypes (AAV2.5 and AAV6) was further evaluated over 8 weeks under varying storage temperatures. Results: Lyophilization in sucrose preserved transduction activity for all tested serotypes (AAV1, AAV2, AAV2.5, AAV3, AAV4, AAV5, AAV6, AAV8, AAV9, and AAVrh10). Notably, AAV2, AAV2.5, and AAV6 exhibited 3- to 6-fold increases in transduction, an effect attributable to the sucrose excipient rather than the lyophilization process itself. Long-term stability of lyophilized vectors varied by serotype, temperature, and vial-seal integrity. AAV6 retained full activity for at least 8 weeks when stored at 4 °C or −20 °C. Conclusions: AAV vectors can be effectively lyophilized in simple sucrose solutions, enabling storage at standard −20 °C freezer temperatures while maintaining functional activity. Optimization of lyophilization buffers and excipients may further extend AAV stability at higher temperatures, improving feasibility for global gene therapy deployment.