Faculty Bibliography

Early diagnosis of Alzheimer's disease plays a pivotal role in patient care and clinical trials. In this study, we have developed a new approach based on 3D deep convolutional neural networks to accurately differentiate mild Alzheimer's disease dementia from mild cognitive impairment and cognitively normal individuals using structural MRIs. For comparison, we have built a reference model based on the volumes and thickness of previously reported brain regions that are known to be implicated in disease progression. We validate both models on an internal held-out cohort from The Alzheimer's Disease Neuroimaging Initiative (ADNI) and on an external independent cohort from The National Alzheimer's Coordinating Center (NACC). The deep-learning model is accurate, achieved an area-under-the-curve (AUC) of 85.12 when distinguishing between cognitive normal subjects and subjects with either MCI or mild Alzheimer's dementia. In the more challenging task of detecting MCI, it achieves an AUC of 62.45. It is also significantly faster than the volume/thickness model in which the volumes and thickness need to be extracted beforehand. The model can also be used to forecast progression: subjects with mild cognitive impairment misclassified as having mild Alzheimer's disease dementia by the model were faster to progress to dementia over time. An analysis of the features learned by the proposed model shows that it relies on a wide range of regions associated with Alzheimer's disease. These findings suggest that deep neural networks can automatically learn to identify imaging biomarkers that are predictive of Alzheimer's disease, and leverage them to achieve accurate early detection of the disease.

Aging-related periventricular white matter hyperintensities (pvWMHs) are a common observation in medical images of the aging brain. The underlying tissue damage is part of the complex pathophysiology associated with age-related microstructural changes and cognitive decline. PvWMH formation is linked to blood-brain barrier dysfunction from cerebral small vessel disease as well as the accumulation of cerebrospinal fluid in periventricular tissue due to progressive denudation of the ventricular wall. In need of a unifying theory for pvWMH etiology, image-based finite-element modeling is used to demonstrate that ventricular expansion from age-related cerebral atrophy and hemodynamic loading leads to maximum mechanical loading of the ventricular wall in the same locations that show pvWMHs. Ventricular inflation, induced via pressurization of the ventricular wall, creates significant ventricular wall stretch and stress on the ependymal cells lining the wall, that are linked to cerebrospinal fluid leaking from the lateral ventricles into periventricular white matter tissue. Eight anatomically accurate 3D brain models of cognitively healthy subjects with a wide range of ventricular shapes are created. For all models, our simulations show that mechanomarkers of mechanical wall loading are consistently highest in pvWMHs locations (p < 0.05). Maximum principal strain, the ependymal cell thinning ratio, and wall curvature are on average 14%, 8%, and 24% higher in pvWMH regions compared to the remaining ventricular wall, respectively. Computational modeling provides a powerful framework to systematically study pvWMH formation and growth with the goal to develop pharmacological interventions in the future.

Given the central role of interstitial fibrosis in disease progression in chronic kidney disease (CKD), a role for diffusion-weighted MRI has been pursued. We evaluated the feasibility and preliminary efficacy of using radiomic features to phenotype apparent diffusion coefficient (ADC) maps and hence to the clinical classification(s) of the participants. The study involved 40 individuals (10 healthy and 30 with CKD (eGFR &lt; 60 mL/min/1.73 m²)). Machine learning methods, such as hierarchical clustering and logistic regression, were used. Clustering resulted in the identification of two clusters, one including all individuals with CKD (n = 17), while the second one included all the healthy volunteers (n = 10) and the remaining individuals with CKD (n = 13), resulting in 100% specificity. Logistic regression identified five radiomic features to classify participants as with CKD vs. healthy volunteers, with a sensitivity and specificity of 93% and 70%, respectively, and an AUC of 0.95. Similarly, four radiomic features were able to classify participants as rapid vs. non-rapid CKD progressors among the 30 individuals with CKD, with a sensitivity and specificity of 71% and 43%, respectively, and an AUC of 0.75. These promising preliminary data should support future studies with larger numbers of participants with varied disease severity and etiologies to improve performance.

BACKGROUND:F-THK5117, we previously reported that the ventricular CSF clearance of the PET tracer was reduced in AD and associated with elevated brain Aβ levels. METHODS:C-PiB to estimate CSF clearance calculated from early dynamic PET frames in 9 normal controls and 15 AD participants. RESULTS:F-THK5351) and brain Aβ load (r =  - 0.64, n = 24, p < 0.01). With a larger sample size, we extended our observations to show that reduced CSF clearance is associated with reductions in cortical thickness and cognitive performance. CONCLUSIONS:Overall, the findings support the hypothesis that failed CSF clearance is a feature of AD that is related to Aβ deposition and to the pathology of AD. Longitudinal studies are needed to determine whether failed CSF clearance is a predictor of progressive amyloidosis or its consequence.

PURPOSE/OBJECTIVE:Gait improvement following high-volume lumbar puncture (HVLP) and continuous lumbar drain (cLD) is widely used to predict shunt response in patients with suspected normal pressure hydrocephalus (NPH). Here, we investigate differences in MRI volumetric and traditional measures between HVLP/cLD responders and non-responders to identify imaging features that may help predict HVLP/cLD response. METHODS:Eighty-two patients with suspected NPH were studied retrospectively. Gait testing was performed before and immediately/24 h/72 h after HVLP/cLD. A positive response was defined as improvement in gait post-procedure. Thirty-six responders (26 men; mean age 79.3 ± 6.3) and 46 non-responders (25 men; mean age 77.2 ± 6.1) underwent pre-procedure brain MRI including a 3D T1-weighted sequence. Subcortical regional volumes were segmented using FreeSurfer. After normalizing for total intracranial volume, two-way type III ANCOVA test and chi-square test were used to characterize statistical group differences. Evans' index, callosal angle (CA), and disproportionately enlarged subarachnoid space hydrocephalus were assessed. Multivariable logistic regression models were tested using Akaike information criterion to determine which combination of metrics most accurately predicts HVLP/cLD response. RESULTS:Responders and non-responders demonstrated no differences in total ventricular and white/gray matter volumes. CA (men only) and third and fourth ventricular volumes were smaller; and hippocampal volume was larger in responders (p < 0.05). Temporal horns volume correlated with degree of improvement in gait velocity in responders (p = 0.0006). The regression model was 76.8% accurate for HVLP/cLD response. CONCLUSION/CONCLUSIONS:CA and third and fourth ventricular volumes and hippocampal volume may serve as potentially useful imaging features that may help predict spinal tap response and hence potentially shunt response.

OBJECTIVE:Texture features are proposed for classification and prognostication, with lacking information about variability. We assessed 3 T liver MRI feature variability. METHODS:Five volunteers underwent standard 3 T MRI, and repeated with identical and altered parameters. Two readers placed regions of interest using 3DSlicer. Repeatability (between standard and repeat scan), robustness (between standard and parameter changed scan), and reproducibility (two reader variation) were computed using coefficient of variation (CV). RESULTS:67%, 49%, and 61% of features had good-to-excellent (CV ≤ 10%) repeatability on ADC, T1, and T2, respectively, least frequently for first order (19-35%). 22%, 19%, and 21% of features had good-to-excellent robustness on ADC, T1, and T2, respectively. 52%, 35%, and 25% of feature measurements had good-to-excellent inter-reader reproducibility on ADC, T1, and T2, respectively, with highest good-to-excellent reproducibility for first order features on ADC/T1. CONCLUSION/CONCLUSIONS:We demonstrated large variations in texture features on 3 T liver MRI. Further study should evaluate methods to reduce variability.

High-resolution susceptibility weighted imaging (SWI) provides unique contrast to small venous vasculature. The conspicuity of these mesoscopic veins, such as deep medullary veins in white matter, is subject to change from SWI venography when venous oxygenation in these veins is altered due to oxygenated blood susceptibility changes. The changes of visualization in small veins shows potential to depict regional changes of oxygen utilization and/or vascular density changes in the aging brain. The goal of this study was to use WM venous density to quantify small vein visibility in WM and investigate its relationship with neurodegenerative features, white matter hyperintensities (WMHs), and cognitive/functional status in elderly subjects (N = 137). WM venous density was significantly associated with neurodegeneration characterized by brain atrophy (β = 0.046± 0.01, p < 0.001), but no significant association was found between WM venous density and WMHs lesion load (p = 0.3963). Further analysis of clinical features revealed a negative trend of WM venous density with the sum-of-boxes of Clinical Dementia Rating and a significant association with category fluency (1-min animal naming). These results suggest that WM venous density on SWI can be used as a sensitive marker to characterize cerebral oxygen metabolism and different stages of cognitive and functional status in neurodegenerative diseases.

Background and objectives/UNASSIGNED:) in a group of cognitively normal middle-aged and older adults. Methods/UNASSIGNED:Our study was a retrospective analysis of prospectively collected data. Subjects were enrolled for studies assessing the role of hippocampal hemodynamics as a biomarker for AD among cognitively healthy elderly individuals (age > 50). Participants without cognitive impairment, stroke, and active substance abuse were recruited between January 2008 and November 2017 at the NYU Grossman School of Medicine, former Center for Brain Health. All subjects underwent medical, psychiatric, and neurological assessments, blood tests, and MRI examinations. To estimate CVR, we increased their carbon dioxide levels using a rebreathing protocol. Relationships between BMI and brain measures were tested using linear regression. Results/UNASSIGNED:in women (β = -0.20, unstandardized B = -0.08, 95% CI -0.13, -0.02). Discussion/UNASSIGNED:These findings lend support to the notion that obesity is a risk factor for hippocampal hemodynamic impairment and suggest targeting obesity as an important prevention strategy. Prospective studies assessing the effects of weight loss on brain hemodynamic measures and inflammation are warranted.

RATIONALE AND OBJECTIVES/OBJECTIVE:Periventricular and deep white matter hyperintensities (WMHs) in the elderly have been reported with distinctive roles in the progression of cognitive decline and dementia. However, the definition of these two subregions of WMHs is arbitrary and varies across studies. Here, we evaluate three partition methods for WMH subregions, including two widely used conventional methods (CV & D10) and one novel method based on bilateral distance (BD). MATERIALS AND METHODS/METHODS:The three partition methods were assessed on the MRI scans of 60 subjects, with 20 normal control, 20 mild cognitive impairment, and 20 Alzheimer's disease (AD). Resulting WMH subregional volumes were (1) compared among different partition methods and subject groups, and (2) tested for clinical associations with cognition and dementia. Inter-rater, intrarater, and interscan reproducibility of WMHs volumes were tested on 12 randomly selected subjects from the 60. RESULTS:For all three partition methods, increased periventricular WMHs were found for AD subjects over normal control. For BD and D10, but not CV method, increased Periventricular WMHs were found for AD subjects over mild cognitive impairment. Significant correlations were found between PVWMHs and Mini-Mental State Examination, Montreal Cognitive Assessment, and Clinical Dementia Rating scores. Furthermore, PVWMHs under BD partition showed higher correlations than D10 and CV. High intrarater and interscan reproducibility (ICCA = 0.998 and 0.992 correspondingly) and substantial inter-rater reproducibility (ICCA = 0.886) were detected. CONCLUSION/CONCLUSIONS:Different WMH partition methods showed comparable diagnostic abilities. The proposed BD method showed advantages in quantifying PVWMH over conventional CV and D10 methods, in terms of higher consistency, larger contrast, and higher diagnosis accuracy. Furthermore, the PVWMH under BD partition showed stronger clinical correlations than conventional methods.

PURPOSE/OBJECTIVE:To assess prognostic value of body composition parameters measured at CT to predict risk of hospitalization in patients with COVID-19 infection. METHODS:177 patients with SARS-CoV-2 infection and with abdominopelvic CT were included in this retrospective IRB approved two-institution study. Patients were stratified based on disease severity as outpatients (no hospital admission) and patients who were hospitalized (inpatients). Two readers blinded to the clinical outcome segmented axial CT images at the L3 vertebral body level for visceral adipose tissue (VAT), subcutaneous adipose tissue (SAT), muscle adipose tissue (MAT), muscle mass (MM). VAT to total adipose tissue ratio (VAT/TAT), MAT/MM ratio, and muscle index (MI) at L3 were computed. These measures, along with detailed clinical risk factors, were compared in patients stratified by severity. Various logistic regression clinical and clinical + imaging models were compared to discriminate inpatients from outpatients. RESULTS:There were 76 outpatients (43%) and 101 inpatients. Male gender (p = 0.013), age (p = 0.0003), hypertension (p = 0.0003), diabetes (p = 0.0001), history of cardiac disease (p = 0.007), VAT/TAT (p < 0.0001), and MAT/MM (p < 0.0001), but not BMI, were associated with hospitalization. A clinical model (age, gender, BMI) had AUC of 0.70. Addition of VAT/TAT to the clinical model improved the AUC to 0.73. Optimal model that included gender, BMI, race (Black), MI, VAT/TAT, as well as interaction between gender and VAT/TAT and gender and MAT/MM demonstrated the highest AUC of 0.83. CONCLUSION/CONCLUSIONS:MAT/MM and VAT/TAT provides important prognostic information in predicting patients with COVID-19 who are likely to require hospitalization.