Citations

Clear

6421 citation(s) found for "Su, Christopher".

Relationships between PET and blood plasma biomarkers in corticobasal syndrome
Alnobani, Alla, Graff‐Radford, Jonathan , Lowe, Val J. , Kanekiyo, Takahisa , Josephs, Keith A. , Whitwell, Jennifer L. , Singh, Neha Atulkumar , Machulda, Mary M. , Mielke, Michelle M. , Schwarz, Christopher G. , Senjem, Matthew L. , Jack, Clifford R.

Mayo Clinic

Alzheimer's & Dementia | January 9, 2025 | Volume: 20 | Issue: S2 | Pages: No page available

10.1002/alz.091767 Get It

AbstractBackgroundCorticobasal syndrome (CBS) is a progressive neurodegenerative syndrome that can result from several underlying neuropathologies, including Alzheimer’s disease (AD). Little is known about the utility of blood plasma measures to predict PET biomarker confirmed AD in CBS.MethodEighteen CBS patients (8 PET Aβ+; 10 PET Aβ‐) and 8 cognitively unimpaired Aβ‐ individuals (CU) were recruited by the Neurodegenerative Research group from the Department of Neurology, Mayo Clinic, Rochester, MN, underwent Aβ (Pittsburgh Compound‐B) and tau (18F‐flortaucipir) PET and provided a blood sample. Blood plasma analysis was performed to assess p‐tau 181, total tau, Aβ 42/40, neurofilament light chain (NfL) and glial fibrillary acidic protein (GFAP) using the Quanterix Simoa platform. Plasma concentrations were compared between Aβ‐, and Aβ+ CBS and controls using Kruskal‐Wallis corrected for multiple comparisons using Dunn’s test, with effect size measured with area under the receiver operating characteristic curve (AUROC) analysis. Spearman correlations were calculated to assess relationships between plasma concentrations and Aβ and tau PET uptake.ResultThe CBS Aβ+ group showed reduced Aβ42/40 ratio, with elevated p‐tau 181, GFAP and NfL plasma concentrations compared to CU, while the CBS Aβ‐ group only showed elevated NfL concentration compared to CU (Figure 1). Both p‐tau 181 (AUROC=0.95) and GFAP (AUROC=0.91) plasma concentrations were able to differentiate CBS Aβ‐ from CBS Aβ+. Plasma p‐tau 181 and GFAP concentrations showed a positive correlation to both Aβ and tau PET uptake, and the Aβ42/40 ratio showed a negative correlation to Aβ PET uptake (Figure 2). No correlations were observed between NfL and Aβ and tau PET.ConclusionThis study supports the use of plasma p‐tau 181 and GFAP to detect AD in CBS, while NfL shows potential as a nonspecific disease biomarker of CBS regardless of underlying pathology. Plasma concentrations of these biomarkers may potentially be incorporated into clinical trials targeting CBS patients with underlying AD pathology.
Aβ42/40 and p‐tau 181 as disease biomarkers in Atypical Alzheimer’s disease
Graff‐Radford, Jonathan , Lowe, Val J. , Josephs, Keith A. , Whitwell, Jennifer L. , Singh, Neha Atulkumar , Li, Danni , Mielke, Michelle M. , Machulda, Mary M. , Schwarz, Christopher G. , Senjem, Matthew L. , Jack, Clifford R.

Mayo Clinic

Alzheimer's & Dementia | January 9, 2025 | Volume: 20 | Issue: S2 | Pages: No page available

10.1002/alz.091571 Get It

AbstractBackgroundBlood plasma metrics provide excellent biomarkers of the presence of Alzheimer’s disease (AD) pathology in population cohorts. However, it is less clear whether plasma metrics correlate with PET measures of AD pathology in cohorts of patients with advanced AD disease, particularly in those with atypical clinical presentations of AD.MethodSeventy‐seven patients (visual variant=43, language variant=32, motor variant=1 and dysexecutive variant=1) with PET biomarker‐confirmed atypical AD were recruited by the Neurodegenerative Research group at Mayo Clinic, Rochester, MN, underwent Aβ (Pittsburgh Compound‐B) and tau (18F‐flortaucipir) PET and provided a blood sample. Blood plasma analysis was performed to assess p‐tau 181 and Aβ42/40 ratio concentrations using the Quanterix Simoa platform. Multivariate linear regression models assessed the relationship between plasma concentrations and both global Aβ, atypical AD summary and regional tau PET measures. For each PET measure, two separate models were fit to assess the influence of each plasma biomarker separately, while the third model included all plasma measures to assess the combined influence of all biomarkers on PET uptake. To investigate differences in plasma concentrations between atypical AD phenotypes, we performed Mann‐Whitney’s test across the visual and language variants of AD.ResultPlasma Aβ42/40 ratio showed positive associations with tau PET uptake, specifically within precuneus, parietal and temporal regions and was not associated with Aβ PET. There was no relationship between plasma p‐tau 181 and Aβ or tau PET uptake. However, the model that included both plasma biomarkers showed that with the addition of p‐tau 181, all significant associations between Aβ42/40 ratio and regional tau uptake were weakened to trends. The Aβ42/40 ratio continued to show no relationship with global Aβ‐PET uptake and plasma p‐tau 181 showed no relationship to PET measures (Figure 1). Lastly, no significant difference in plasma Aβ42/40 and p‐tau 181 levels was noted between the visual and language variants of AD (Figure 2).ConclusionThe lack of associations between plasma concentrations and PET measures of Aβ and tau in this relatively advanced AD cohort may suggest that plasma concentrations have greater utility as diagnostic biomarkers of early AD as compared to advanced AD.
Comparing de‐facing software MiDeFace and mri_reface according to privacy protection and effects on downstream measurements
Prakaashana, Carl M., Gunter, Jeffrey L. , Graff‐Radford, Jonathan, Petersen, Ronald C. , Initiative, The Alzheimer's Disease Neuroimaging, Schwarz, Christopher G. , Kremers, Walter K. , Senjem, Matthew L. , Vemuri, Prashanthi , Kantarci, Kejal , Graff‐Radford, Jonathan , Knopman, David S. , Jack, Clifford R. , Initiative, The Alzheimer's Disease Neuroimaging

Mayo Clinic

Alzheimer's & Dementia | January 9, 2025 | Volume: 20 | Issue: S2 | Pages: No page available

10.1002/alz.091376 Get It

AbstractBackgroundSeveral programs are available for automatically “de‐facing” brain images. These are designed to prevent the potential use of face imagery to re‐identify research participants. Previous works have compared available software and mri_reface is frequently among the top methods. The “Minimally Invasive DeFacing of MRI images” (MiDeFace) program was more recently added to FreeSurfer, but minimal validation of MiDeFace has been published.MethodWe ran mri_reface 0.3.3 and MiDeFace 7.3.2 on two datasets: A) 207 participants with face photographs and 3D T1‐weighted and 3D T2‐weighted‐FLAIR MRI from the Mayo Clinic Study of Aging (all unimpaired, Siemens); and B) 300 participants from ADNI3 with 3D T1‐weighted MRI (50 unimpaired + 50 with clinical AD, from each of 3 vendors). For dataset A, we tested the rate at which Microsoft Azure face recognition matched each participant’s photographs with the correct MRI, comparing unmodified images, those from mri_reface, and those from MiDeFace. For dataset B, we compared measured regional brain volume and cortical thickness from original vs. de‐faced images with SPM12, FreeSurfer 7.4.1, and the FSL‐UK Biobank pipeline. For both datasets, images were visually rated to assess removal of the face and retention of brain tissue.ResultAutomated face recognition match rates were lower with mri_reface (17/207 8.2%) than with MiDeFace (24/207 11.6%), indicating stronger identity protection with mri_reface, but both were greatly reduced vs. unmodified images (203/207 98.1%). Example images with both programs are shown in Figure 1. From visual ratings (Table 1), rates of complete success (modification of the entire face while retaining the entire brain) were much higher with mri_reface, but the great majority of failures from MiDeFace were subtle and occurred in relatively less recognizable face regions. The effects of both programs on downstream imaging measurements (Figure 2) were small and very comparable, but MiDeFace was favored mildly for SPM12 and for FSL, while mri_reface was more favored for FreeSurfer.ConclusionAltogether, mri_reface more reliably removed the complete face, leading to smaller rates of potential re‐identification (8.2% vs 11.6%). Effects on downstream image measurements were small and comparable with both methods.
A Large Public Release of Clinical and Imaging Data from the Mayo Clinic Study of Aging
Prakaashana, Carl M., Przybelski, Scott A. , Christenson, Luke R., Wiliams, Josie M., Gunter, Jeffrey L. , Fields, Julie A. , Lowe, Val J. , Graff‐Radford, Jonathan , Petersen, Ronald C. , Schwarz, Christopher G. , Kremers, Walter K. , Senjem, Matthew L. , Arani, Arvin , Reid, Robert I. , Machulda, Mary M. , Kantarci, Kejal , Vemuri, Prashanthi , Knopman, David S. , Jack, Clifford R.

Mayo Clinic

Alzheimer's & Dementia | January 9, 2025 | Volume: 20 | Issue: S2 | Pages: No page available

10.1002/alz.091346 Get It

Abstract Background The Mayo Clinic Study of Aging (MCSA) is a longitudinal, population‐based study of residents of Olmsted County, Minnesota, USA. MCSA is releasing de‐identified clinical and imaging data on GAAIN.org to benefit the research community. Method We included longitudinal clinical data from all participants of the Mayo Clinic Study of Aging 30‐90 years of age (average 70.0y) through the first 15 years of the study. Clinical data includes age, sex, self‐reported race/ethnicity, APoE4 allele status, cognitive impairment status (normal, MCI, dementia, or other), height, weight, blood pressure, multi‐domain neuropsychological scores, neuroimaging summary measurements, medications, cardiovascular risk factors, social activity, sleep, mood/anxiety, functional measures, and more. For our initial release of imaging data, we selected the first imaging timepoint acquired on 3T GE MRI scanners and amyloid (PiB) PET during the same visit, from all available imaging participants 50‐90 years of age (average 70.7y). Currently, we included 3D T1‐weighted (MPRAGE) MRI (1x1x1.2mm 3 ), T2‐weighted FLAIR MRI (0.85x0.85x3.6mm 3 ), diffusion MRI (2.7mm resolution with 41 directions at b=1000 and 5 b=0), and numeric cortical global SUVR and Centiloid values from PiB PET scans at 40‐60 minutes post‐injection. All images are released in DICOM format and thoroughly de‐identified using CTP while ensuring preservation of necessary fields for analyses. T1‐weighted and T2‐FLAIR MRI were also de‐faced (identifiable face imagery was removed and replaced with an average) using mri_reface version 0.3.3 to further protect participants’ privacy, and all de‐faced images were visually inspected by a trained data scientist for quality assurance. Result The first phase of our public data release includes clinical data from 5925 unique participants from the Mayo Clinic Study of Aging (MCSA), each ranging from 1‐12 visits at roughly 15‐month intervals, and three MRI sequences (T1‐w, T2‐w‐FLAIR, and diffusion) from 1802 unique participants. Demographic summaries are given in Table 1. Conclusion This large, richly characterized, de‐identified clinical and imaging dataset is available to the research community at https://www.gaaindata.org/partner/MCSA , and it is suitable for a wide range of research uses. Additional data will be released over time, including de‐faced amyloid PET images.
Distinct Spatial Patterns for White Matter Microstructure Changes in an Aging and Alzheimer’s Disease Cohort
Tian, Jianqiao , Raghavan, Sheelakumari , Przybelski, Scott A. , Graff‐Radford, Jonathan, Lowe, Val J. , Petersen, Ronald C. , Cogswell, Petrice M. , Reid, Robert I. , Schwarz, Christopher G. , Graff‐Radford, Jonathan , Kantarci, Kejal , Knopman, David S. , Jack, Clifford R. , Vemuri, Prashanthi

Mayo Clinic

Alzheimer's & Dementia | January 9, 2025 | Volume: 20 | Issue: S2 | Pages: No page available

10.1002/alz.090927 Get It

AbstractBackgroundWhite matter (WM) damage is seen with neurodegenerative and cerebrovascular pathologies and contributes to cognitive dysfunction. We hypothesized that there are multiple disease progression patterns in WM microstructural changes related to aging and dementia, and these can be identified using SuStaIn (data‐driven clustering algorithm for disease subtype and stage discovery) on multishell diffusion MRI (NODDI) data. NODDI provides information on cellular tissue architecture – neurite density index (NDI) measures packing density of axons and dendrites, and free water fraction (FWF) measures unrestricted water, e.g., interstitial and cerebrospinal fluid.MethodWe identified 1136 participants (mean age=70 years, 51% male, 40% Aβ‐positive, 19% MCI/AD dementia) from the Mayo ADRC and Mayo Clinic Study of Aging. We analyzed NDI and FWF in 55 WM regions. We selected a subset of WM features that monotonically changed with clinical disease progression and then applied the Z‐score SuStaIn algorithm.ResultNDI in seven regions and FWF in six regions survived selection criteria (Figure 1). Noteworthy, the NDI regions were predominantly located in the temporal lobe, while FWF regions were primarily distributed in the frontal and subcortical regions. The SuStaIn algorithm discovered two stable clusters: the NDI‐First (Temporal) and FWF‐First (Frontal) groups (Figure 2). For the NDI‐First group, temporal lobe NDI changed early followed by frontal FWF. The FWF‐First group had the reverse order. There was clear phase separation based on which measure changed first. Stages of the NDI‐First group correlated better with meta‐ROI Tau‐PET SUVR and with MMSE than the FWF‐First group (p<0.01), suggesting that NDI‐First may be indicative of a primarily AD pathway.ConclusionGiven NDI changes are likely degenerative and FWF are suggested to be other processes such as neurodegeneration or inflammatory, the observed changes in the temporal and frontal regions may be due to distinct etiologies, i.e., neurodegenerative in the temporal lobes vs. vascular dysfunction often observed in the frontal lobes. The SuStaIn algorithm showed promise in discovering disease subtypes of WM microstructural changes. Further work is needed to understand the mechanistic basis of these observed changes and integrate WM changes for differential etiology identification and disease staging.
Cerebrovascular reactivity (CVR) MRI as a biomarker for cerebral small vessel disease (SVD) related cognitive decline: Multi‐site validation in the MarkVCID Consortium
Hazel, Kaisha , Pottanat, George , Xu, Cuimei, Lucke, Emma, Gold, Brian T. , Greenberg, Steven M. , Helmer, Karl G. , Kramer, Joel H. , Mulavelil, Rachel, Singh, Herpreet , Velarde, Angel G., Kalyani, Rita R. , Rosenberg, Paul B. , Albert, Marilyn , Liu, Peiying , Lin, Zixuan , Jiang, Dengrong , Pillai, Jay J. , Bauer, Christopher E. , Jann, Kay , Jicha, Gregory A. , Maillard, Pauline , Satizábal, Claudia L. , Schwab, Kristin , Seshadri, Sudha , Wang, Danny J.J. , Moghekar, Abhay , Yaşar, Sevil , Lu, Hanzhang

Mayo Clinic

Alzheimer's & Dementia | January 9, 2025 | Volume: 20 | Issue: S2 | Pages: No page available

10.1002/alz.089224 Get It

AbstractBackgroundCerebral small vessel disease (SVD) related vascular contributions represent a major factor contributing to cognitive decline and dementia (VCID) in older adults. However, there has not been a validated biomarker for the diagnosis and treatment monitoring of this condition. Recently, the US National Institute on Neurological Disorders and Stroke (NINDS) funded the MarkVCID Consortium to identify and validate clinical‐trial‐ready biomarkers for VCID. Cerebrovascular reactivity (CVR) MRI is one of the selected biomarkers that underwent multi‐site testing in the Consortium. The present study aimed to report the relationship between CVR and cognitive function at independent sites, based on a pre‐specified hypothesis.MethodCVR, the ability of cerebral small vessels to dilate upon stimulus, is thought to directly reflect physiological function of the brain microvasculature. Based on previous single‐site findings, the pre‐defined hypothesis was that CVR will be associated with the global cognitive function measured by Montreal Cognitive Assessment (MoCA) after adjusting for age, sex, and education, and this association will be observed in data collected and analyzed at each individual site.A total of 264 older participants from three sites were included (Table 1). Each site performed an identical CVR MRI procedure using 5% CO2 inhalation, and a standardized clinical and cognitive evaluation. Multi‐linear regression analysis was conducted on a site‐by‐site basis to examine the pre‐defined hypothesis.ResultGray‐matter CVR showed a positive association with MoCA score after adjustment for age, sex, and education, in which participants with higher CVR had higher MoCA scores. This relationship was reproduced at each site (Figure 1, p<0.05 for each), confirming our pre‐specified hypothesis.In the secondary analysis of all data together, higher gray‐matter CVR was found to be significantly associated with better executive function measure of item response theory (IRT) score (β=2.87, p=0.003).ConclusionThe present study evaluated the relationship between CVR and cognition in a multi‐center setting. CVR was found to be positively associated with global cognitive function measured by the MoCA, which was shown to be reproducible across different sites with diverse cohorts. These findings support the utility of CVR as a biomarker in future clinical trials of SVD and VCID.
Outcomes of Immunotherapy Treatment in Sinonasal Mucosal Melanoma
Kshirsagar, Rijul S. , Harris, Jacob , Chang, Eugene H. , Fung, Nicholas , Hong, Michelle , Johnson, Brian J. , Kohanski, Michael A. , Lee, Jivianne T. , Nabavizadeh, Seyed Ali , Obermeyer, Isaac P. , Pandrangi, Vivek C. , Smith, Timothy L. , Wang, Marilene B. , Lazor, Jillian W. , Mitchell, Tara C. , Adappa, Nithin D. , Adappa, Nithin D. , Harris, Jacob , Lee, Jivianne T , Obermeyer, Isaac P , Smith, Timothy L , Snyderman, Carl H , Kuan, Edward C , Kshirsagar, Rijul S , Eide, Jacob G. , Abiri, Arash , Beswick, Daniel M. , Fung, Nicholas, Le, Christopher H. , Pinheiro‐Neto, Carlos , Suh, Jeffrey D. , Wang, Eric W , Choby, Garret , Geltzeiler, Mathew , Adappa, Nithin D.

Mayo Clinic

American Journal of Rhinology & Allergy | January 9, 2025 | Volume: 39 | Issue: 2 | Pages: 102-108

10.1177/19458924241308953 | PMID 39782303 Get It

Background Sinonasal mucosal melanoma has poor survival despite multimodality treatment. While the impact of immunotherapy (IT) on metastatic cutaneous melanoma is well-defined, there are relatively little data on sinonasal mucosal melanoma. Objective We sought to define immunotherapy outcomes in patients with sinonasal mucosal melanoma. Methods A retrospective cohort study evaluated patients treated with IT during their overall treatment strategy for SNMM. Patient demographics, treatment, and survival outcomes were recorded. Results 52 patients had IT treatment for SNMM from 2000 to 2022, with an average age of 69.1 ± 11.9 years. The most common treatment was surgery with radiation and IT ( n = 26, 50%). Most regimens consisted of a combination of Nivolumab and Ipilimumab ( n = 17, 32.7%) or pembrolizumab ( n = 14, 26.9%). 44.2% of patients experienced reported complications. Overall survival at 1-, 2-, and 5 years was 86.9%, 74.1%, and 39.1%, respectively. Conclusion Approximately half of patients will have a local response following immunotherapy, but it is rare to have improvement at metastatic locations. Further research within our group will assess optimal timing and markers that are predictive of response.
Relationships between PET and blood plasma biomarkers in corticobasal syndrome
Alnobani, Alla, Graff‐Radford, Jonathan, Schwarz, Christopher G., Lowe, Val J. , Kanekiyo, Takahisa , Josephs, Keith A. , Whitwell, Jennifer L. , Singh, Neha Atulkumar , Graff‐Radford, Jonathan , Machulda, Mary M. , Mielke, Michelle M. , Schwarz, Christopher G. , Senjem, Matthew L. , Jack, Clifford R.

Mayo Clinic

Alzheimer's & Dementia | January 9, 2025 | Volume: 20 | Issue: S9 | Pages: No page available

10.1002/alz.094001 Get It

AbstractBackgroundCorticobasal syndrome (CBS) is a progressive neurodegenerative syndrome that can result from several underlying neuropathologies, including Alzheimer’s disease (AD). Little is known about the utility of blood plasma measures to predict PET biomarker confirmed AD in CBS.MethodEighteen CBS patients (8 PET Aß+; 10 PET Aß‐) and 8 cognitively unimpaired Aß‐ individuals (CU) were recruited by the Neurodegenerative Research group from the Department of Neurology, Mayo Clinic, Rochester, MN, underwent Aß (Pittsburgh Compound‐B) and tau (18F‐flortaucipir) PET and provided a blood sample. Blood plasma analysis was performed to assess p‐tau 181, total tau, Aß 42/40, neurofilament light chain (NfL) and glial fibrillary acidic protein (GFAP) using the Quanterix Simoa platform. Plasma concentrations were compared between Aß‐, and Aß+ CBS and controls using Kruskal‐Wallis corrected for multiple comparisons using Dunn’s test, with effect size measured with area under the receiver operating characteristic curve (AUROC) analysis. Spearman correlations were calculated to assess relationships between plasma concentrations and Aß and tau PET uptake.ResultThe CBS Aß+ group showed reduced Aß42/40 ratio, with elevated p‐tau 181, GFAP and NfL plasma concentrations compared to CU, while the CBS Aß‐ group only showed elevated NfL concentration compared to CU (Figure 1). Both p‐tau 181 (AUROC=0.95) and GFAP (AUROC=0.91) plasma concentrations were able to differentiate CBS Aß‐ from CBS Aß+. Plasma p‐tau 181 and GFAP concentrations showed a positive correlation to both Aß and tau PET uptake, and the Aß42/40 ratio showed a negative correlation to Aß PET uptake (Figure 2). No correlations were observed between NfL and Aß and tau PET.ConclusionThis study supports the use of plasma p‐tau 181 and GFAP to detect AD in CBS, while NfL shows potential as a nonspecific disease biomarker of CBS regardless of underlying pathology. Plasma concentrations of these biomarkers may potentially be incorporated into clinical trials targeting CBS patients with underlying AD pathology.
Measuring the Potential Risk of Re‐identification of Imaging Research Participants from Open‐Source Automated Face Recognition Software
Prakaashana, Carl M., Savvides, Marios , Gunter, Jeffrey L. , Graff‐Radford, Jonathan, Petersen, Ronald C. , Senjem, Matthew L. , Vemuri, Prashanthi , Kantarci, Kejal , Graff‐Radford, Jonathan , Knopman, David S. , Jack, Clifford R. , Schwarz, Christopher G.

Mayo Clinic

Alzheimer's & Dementia | January 9, 2025 | Volume: 20 | Issue: S9 | Pages: No page available

10.1002/alz.093977 Get It

AbstractBackgroundResearch subjects can potentially be re‐identified from de‐identified MRI, CT, and PET brain scans with up to 98% accuracy using Microsoft Azure’s cloud‐based commercial facial recognition software. This showed the need to “de‐face” publicly shared research brain scans. Subsequently, Microsoft has begun restricting its face recognition services, intending to prevent misuse. This study tests a variety of popular open‐source computer vision and facial recognition software packages to measure the re‐identification risk with only unrestricted free software.MethodThis study used brain MRI from Siemens 3T scanners and face photographs from 182 participants (ages 30‐90+) in the Mayo Clinic Study of Aging. Each participant had photos taken with an iPad, which were cropped loosely around the head and converted to black and white. A total of 81 renders (2D reconstruction images; Figure 1) were made from each 3D T2‐FLAIR MRI, simulating various camera and lighting positions. These face renders approximate black and white photos, allowing use with software packages developed for matching faces/features between photographs. Open‐source Python packages (OpenCV, DeepFace, face_recognition, scikit‐learn) and their included pre‐trained models (designed for photographs) were used to match participant photographs to the labelled library of MRI‐based face renders. A correct match was recorded when the top‐scoring render or set of renders automatically chosen by the algorithm was from the MRI of the correct participant.ResultFeature matching with ORB and SWIFT (both from OpenCV) produced no correct matches, despite a prior publication reporting moderate success. The best match rates (Figure 2) were 58‐59% accuracy using face_recognition or DeepFace packages with support vector machines (SVMs). Without SVMs, these packages performed modestly worse (51%‐55%).ConclusionWhile the highest match rates obtained by widely available open‐source packages (roughly 60%) are much lower than those of a limited‐access commercial product (Microsoft Azure, with 98%), they show that actors with only access to no‐cost readily available software could still re‐identify a research subject from a brain scan most of the time. This further highlights the need for image refacing software, such as mri_reface, to protect the privacy and identifies of research volunteers.
Comparing de‐facing software MiDeFace and mri_reface according to privacy protection and effects on downstream measurements
Prakaashana, Carl M., Gunter, Jeffrey L. , Graff‐Radford, Jonathan, Petersen, Ronald C. , Schwarz, Christopher G. , Kremers, Walter K. , Senjem, Matthew L. , Vemuri, Prashanthi , Kantarci, Kejal , Graff‐Radford, Jonathan , Knopman, David S. , Jack, Clifford R.

Mayo Clinic

Alzheimer's & Dementia | January 9, 2025 | Volume: 20 | Issue: S9 | Pages: No page available

10.1002/alz.093968 Get It

AbstractBackgroundSeveral programs are available for automatically “de‐facing” brain images. These are designed to prevent the potential use of face imagery to re‐identify research participants. Previous works have compared available software and mri_reface is frequently among the top methods. The “Minimally Invasive DeFacing of MRI images” (MiDeFace) program was more recently added to FreeSurfer, but minimal validation of MiDeFace has been published.MethodWe ran mri_reface 0.3.3 and MiDeFace 7.3.2 on two datasets: A) 207 participants with face photographs and 3D T1‐weighted and 3D T2‐weighted‐FLAIR MRI from the Mayo Clinic Study of Aging (all unimpaired, Siemens); and B) 300 participants from ADNI3 with 3D T1‐weighted MRI (50 unimpaired + 50 with clinical AD, from each of 3 vendors). For dataset A, we tested the rate at which Microsoft Azure face recognition matched each participant’s photographs with the correct MRI, comparing unmodified images, those from mri_reface, and those from MiDeFace. For dataset B, we compared measured regional brain volume and cortical thickness from original vs. de‐faced images with SPM12, FreeSurfer 7.4.1, and the FSL‐UK Biobank pipeline. For both datasets, images were visually rated to assess removal of the face and retention of brain tissue.ResultAutomated face recognition match rates were lower with mri_reface (17/207 8.2%) than with MiDeFace (24/207 11.6%), indicating stronger identity protection with mri_reface, but both were greatly reduced vs. unmodified images (203/207 98.1%). Example images with both programs are shown in Figure 1. From visual ratings (Table 1), rates of complete success (modification of the entire face while retaining the entire brain) were much higher with mri_reface, but the great majority of failures from MiDeFace were subtle and occurred in relatively less recognizable face regions. The effects of both programs on downstream imaging measurements (Figure 2) were small and very comparable, but MiDeFace was favored mildly for SPM12 and for FSL, while mri_reface was more favored for FreeSurfer.ConclusionAltogether, mri_reface more reliably removed the complete face, leading to smaller rates of potential re‐identification (8.2% vs 11.6%). Effects on downstream image measurements were small and comparable with both methods.