RESUMO
Objective: Multimorbidity burden across disease cohorts and variations in clinico-radiographic presentations within normal pressure hydrocephalus (NPH) confound its diagnosis, and the assessment of its amenability to interventions. We hypothesized that novel imaging techniques such as 3-directional linear morphological indices could help in distinguishing between hydrocephalus vs. non-hydrocephalus and correlate with responsiveness to external lumbar drainage (CSF responsiveness) within NPH subtypes. Methodology: Twenty-one participants with NPH were recruited and age-matched to 21 patients with Alzheimer's Disease (AD) and 21 healthy controls (HC) selected from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. Patients with NPH underwent testing via the NPH programme with external lumbar drainage (ELD); pre- and post-ELD MRI scans were obtained. The modified Frailty Index (mFI-11) was used to stratify the NPH cohort, including Classic and Complex subtypes, by their comorbidity and frailty risks. The quantitative imaging network tool 3D Slicer was used to derive traditional 2-dimensional (2d) linear measures; Evans Index (EI), Bicaudate Index (BCI) and Callosal Angle (CA), along with novel 3-directional (3d) linear measures; z-Evans Index and Brain per Ventricle Ratio (BVR). 3-Dimensional (3D) ventricular volumetry was performed as an independent correlate of ventriculomegaly to CSF responsiveness. Results: Mean age for study participants was 71.14 ± 6.3 years (18, 85.7% males). The majority (15/21, 71.4%) of participants with NPH comprised the Complex subtype (overlay from vascular risk burden and AD); 12/21 (57.1%) were Non-Responders to ELD. Frailty alone was insufficient in distinguishing between NPH subtypes. By contrast, 3d linear measures distinguished NPH from both AD and HC cohorts, but also correlated to CSF responsiveness. The z-Evans Index was the most sensitive volumetric measure of CSF responsiveness (p = 0.012). Changes in 3d morphological indices across timepoints distinguished between Responders vs. Non-Responders to lumbar testing. There was a significant reduction of indices, only in Non-Responders and across multiple measures (z-Evans Index; p = 0.001, BVR at PC; p = 0.024). This was due to a significant decrease in ventricular measurement (p = 0.005) that correlated to independent 3D volumetry (p = 0.008). Conclusion. In the context of multimorbidity burden, frailty risks and overlay from neurodegenerative disease, 3d morphological indices demonstrated utility in distinguishing hydrocephalus vs. non-hydrocephalus and degree of CSF responsiveness. Further work may support the characterization of patients with Complex NPH who would best benefit from the risks of interventions.
RESUMO
AIM: Attenuation correction using zero-echo time (ZTE) - magnetic resonance imaging (MRI) (ZTE-MRAC) has become one of the standard methods for brain-positron emission tomography (PET) on commercial PET/MR scanners. Although the accuracy of the net tracer-uptake quantification based on ZTE-MRAC has been validated, that of the diagnosis for dementia has not yet been clarified, especially in terms of automated statistical analysis. The aim of this study was to clarify the impact of ZTE-MRAC on the diagnosis of Alzheimer's disease (AD) by performing simulation study. METHODS: We recruited 27 subjects, who underwent both PET/computed tomography (CT) and PET/MR (GE SIGNA) examinations. Additionally, we extracted 107 subjects from the Alzheimer Disease Neuroimaging Initiative (ADNI) dataset. From the PET raw data acquired on PET/MR, three FDG-PET series were generated, using two vendor-provided MRAC methods (ZTE and Atlas) and CT-based AC. Following spatial normalization to Montreal Neurological Institute (MNI) space, we calculated each patient's specific error maps, which correspond to the difference between the PET image corrected using the CTAC method and the PET images corrected using the MRAC methods. To simulate PET maps as if ADNI data had been corrected using MRAC methods, we multiplied each of these 27 error maps with each of the 107 ADNI cases in MNI space. To evaluate the probability of AD in each resulting image, we calculated a cumulative t-value using a fully automated method which had been validated not only in the original ADNI dataset but several multi-center studies. In the method, PET score = 1 is the 95% prediction limit of AD. PET score and diagnostic accuracy for the discrimination of AD were evaluated in simulated images using the original ADNI dataset as reference. RESULTS: Positron emission tomography score was slightly underestimated both in ZTE and Atlas group compared with reference CTAC (-0.0796 ± 0.0938 vs. -0.0784 ± 0.1724). The absolute error of PET score was lower in ZTE than Atlas group (0.098 ± 0.075 vs. 0.145 ± 0.122, p < 0.001). A higher correlation to the original PET score was observed in ZTE vs. Atlas group (R 2: 0.982 vs. 0.961). The accuracy for the discrimination of AD patients from normal control was maintained in ZTE and Atlas compared to CTAC (ZTE vs. Atlas. vs. original; 82.5% vs. 82.1% vs. 83.2% (CI 81.8-84.5%), respectively). CONCLUSION: For FDG-PET images on PET/MR, attenuation correction using ZTE-MRI had superior accuracy to an atlas-based method in classification for dementia. ZTE maintains the diagnostic accuracy for AD.
RESUMO
Mild cognitive impairment (MCI) is a prodromal stage of dementia.Predicting MCI's conversion to Alzheimer's disease (AD) plays critical roles in preventing the progression of AD.Alzheimer's disease neuroimaging initiative (ADNI) was introduced briefly,which was a widely used neuroimaging database for the study on AD related diseases,and the application of machine learning algorithm was reviewed in MCI classification.Deep learning network,which transforms the original data into a higher level and more abstract expression,has shown great promise in MCI conversion and classification.Two main kinds of deep learning approaches were described,including supervised learning and unsupervised learning,and their new application was discussed in MCI conversion and classification based on structural magnetic resonance imaging (sMRI).Finally,the current limitations and future trends of deep learning in this area were explored.
RESUMO
Mild cognitive impairment (MCI) is a prodromal stage of dementia.Predicting MCI's conversion to Alzheimer's disease (AD) plays critical roles in preventing the progression of AD.Alzheimer's disease neuroimaging initiative (ADNI) was introduced briefly,which was a widely used neuroimaging database for the study on AD related diseases,and the application of machine learning algorithm was reviewed in MCI classification.Deep learning network,which transforms the original data into a higher level and more abstract expression,has shown great promise in MCI conversion and classification.Two main kinds of deep learning approaches were described,including supervised learning and unsupervised learning,and their new application was discussed in MCI conversion and classification based on structural magnetic resonance imaging (sMRI).Finally,the current limitations and future trends of deep learning in this area were explored.
