Predictors of multiple sclerosis progression: A systematic review of conventional magnetic resonance imaging studies.

INTRODUCTION: Multiple Sclerosis (MS) is a chronic neurodegenerative disorder that affects the central nervous system (CNS) and results in progressive clinical disability and cognitive decline. Currently, there are no specific imaging parameters available for the prediction of longitudinal disability in MS patients. Magnetic resonance imaging (MRI) has linked imaging anomalies to clinical and cognitive deficits in MS. In this study, we aimed to evaluate the effectiveness of MRI in predicting disability, clinical progression, and cognitive decline in MS. METHODS: In this study, according to PRISMA guidelines, we comprehensively searched the Web of Science, PubMed, and Embase databases to identify pertinent articles that employed conventional MRI in the context of Relapsing-Remitting and progressive forms of MS. Following a rigorous screening process, studies that met the predefined inclusion criteria were selected for data extraction and evaluated for potential sources of bias. RESULTS: A total of 3028 records were retrieved from database searching. After a rigorous screening, 53 records met the criteria and were included in this study. Lesions and alterations in CNS structures like white matter, gray matter, corpus callosum, thalamus, and spinal cord, may be used to anticipate disability progression. Several prognostic factors associated with the progression of MS, including presence of cortical lesions, changes in gray matter volume, whole brain atrophy, the corpus callosum index, alterations in thalamic volume, and lesions or alterations in cross-sectional area of the spinal cord. For cognitive impairment in MS patients, reliable predictors include cortical gray matter volume, brain atrophy, lesion characteristics (T2-lesion load, temporal, frontal, and cerebellar lesions), white matter lesion volume, thalamic volume, and corpus callosum density. CONCLUSION: This study indicates that MRI can be used to predict the cognitive decline, disability progression, and disease progression in MS patients over time.

Early cortical atrophy is related to depression in patients with neuropathologically confirmed Parkinson's disease.

OBJECTIVE: Depression is a common comorbidity in Parkinson's disease (PD) and other synucleinopathies. In non-PD geriatric patients, cortical atrophy has previously been connected to depression. Here, we investigated cortical atrophy and vascular white matter hyperintensities (WMHs) in autopsy-confirmed parkinsonism patients with the focus on clinical depression. METHODS: The sample consisted of 50 patients with a postmortem confirmed neuropathological diagnosis (30 Parkinson's disease [PD], 10 progressive supranuclear palsy [PSP] and 10 multiple system atrophy [MSA]). Each patient had been scanned with brain computerized tomography (CT) antemortem (median motor symptom duration at scanning = 3.0 years), and 19 patients were scanned again after a mean interval of 2.7 years. Medial temporal atrophy (MTA), global cortical atrophy (GCA) and WMHs were evaluated computationally from CT scans using an image quantification tool based on convolutional neural networks. Depression and other clinical parameters were recorded from patient files. RESULTS: Depression was associated with increased MTA after controlling for diagnosis, age, symptom duration, and cognition (p = 0.006). A similar finding was observed with GCA (p = 0.017) but not with WMH (p = 0.47). In PD patients alone, the result was confirmed for MTA (p = 0.021) with the same covariates. In the longitudinal analysis, GCA change per year was more severe in depressed patients than in nondepressed patients (p = 0.029). CONCLUSIONS: Early medial temporal and global cortical atrophy, as detected with automated analysis of CT-images using convolutional neural networks, is associated with clinical depression in parkinsonism patients. Global cortical atrophy seems to progress faster in depressed patients.

Predicting amyloid-PET status in a memory clinic: The role of the novel antero-posterior index and visual rating scales.

INTRODUCTION: Visual rating scales are increasingly utilized in clinical practice to assess atrophy in crucial brain regions among patients with cognitive disorders. However, their capacity to predict Alzheimer's disease (AD)-related pathology remains unexplored, particularly within a heterogeneous memory clinic population. This study aims to assess the accuracy of a novel visual rating assessment, the antero-posterior index (API) scale, in predicting amyloid-PET status. Furthermore, the study seeks to determine the optimal cohort-based cutoffs for the medial temporal atrophy (MTA) and parietal atrophy (PA) scales and to integrate the main visual rating scores into a predictive model. METHODS: We conducted a retrospective analysis of brain MRI and high-resolution TC scans from 153 patients with cognitive disorders who had undergone amyloid-PET assessments due to suspected AD pathology in a real-world memory clinic setting. RESULTS: The API scale (cutoff ≥1) exhibited the highest accuracy (AUC = 0.721) among the visual rating scales. The combination of the cohort-based MTA and PA threshold with the API yielded favorable accuracy (AUC = 0.787). Analyzing a cohort of MCI/Mild dementia patients below 75 years of age, the API scale and the predictive model improved their accuracy (AUC = 0.741 and 0.813, respectively), achieving excellent results in the early-onset population (AUC = 0.857 and 0.949, respectively). CONCLUSION: Our study emphasizes the significance of visual rating scales in predicting amyloid-PET positivity within a real-world memory clinic. Implementing the novel API scale, alongside our cohort-based MTA and PA thresholds, has the potential to substantially enhance diagnostic accuracy.

High throughput method for detecting murine brain atrophy using a clinical 3T MRI.

BACKGROUND: There is a lack of understanding of the mechanisms by which the CNS is injured in multiple sclerosis (MS). Since Theiler's murine encephalomyelitis virus (TMEV) infection in SJL/J mice is an established model of progressive disability in MS, and CNS atrophy correlates with progressive disability in MS, we used in vivo MRI to quantify total ventricular volume in TMEV infection. We then sought to identify immunological and virological biomarkers that correlated with increased ventricular size. METHODS: Mice, both infected and control, were followed for 6 months. Cerebral ventricular volumes were determined by MRI, and disability was assessed by Rotarod. A range of immunological and virological measures was obtained using standard techniques. RESULTS: Disability was present in infected mice with enlarged ventricles, while infected mice without enlarged ventricles had Rotarod performance similar to sham mice. Ventricular enlargement was detected as soon as 1 month after infection. None of the immunological and virological measures correlated with the development of ventricular enlargement. CONCLUSIONS: These results support TMEV infection with brain MRI monitoring as a useful model for exploring the biology of disability progression in MS, but they did not identify an immunological or virological correlate with ventricular enlargement.

Preliminary validation of a structural magnetic resonance imaging metric for tracking dementia-related neurodegeneration and future decline.

Alzheimer's disease (AD) is a progressive neurodegenerative disease characterized by cognitive decline and atrophy in the medial temporal lobe (MTL) and subsequent brain regions. Structural magnetic resonance imaging (sMRI) has been widely used in research and clinical care for diagnosis and monitoring AD progression. However, atrophy patterns are complex and vary by patient. To address this issue, researchers have made efforts to develop more concise metrics that can summarize AD-specific atrophy. Many of these methods can be difficult to interpret clinically, hampering adoption. In this study, we introduce a novel index which we call an "AD-NeuroScore," that uses a modified Euclidean-inspired distance function to calculate differences between regional brain volumes associated with cognitive decline. The index is adjusted for intracranial volume (ICV), age, sex, and scanner model. We validated AD-NeuroScore using 929 older adults from the Alzheimer's Disease Neuroimaging Initiative (ADNI) study, with a mean age of 72.7 years (SD = 6.3; 55.1-91.5) and cognitively normal (CN), mild cognitive impairment (MCI), or AD diagnoses. Our validation results showed that AD-NeuroScore was significantly associated with diagnosis and disease severity scores (measured by MMSE, CDR-SB, and ADAS-11) at baseline. Furthermore, baseline AD-NeuroScore was associated with both changes in diagnosis and disease severity scores at all time points with available data. The performance of AD-NeuroScore was equivalent or superior to adjusted hippocampal volume (AHV), a widely used metric in AD research. Further, AD-NeuroScore typically performed as well as or sometimes better when compared to other existing sMRI-based metrics. In conclusion, we have introduced a new metric, AD-NeuroScore, which shows promising results in detecting AD, benchmarking disease severity, and predicting disease progression. AD-NeuroScore differentiates itself from other metrics by being clinically practical and interpretable.

Brain volume loss in multiple sclerosis is independent of disease activity and might be prevented by early disease-modifying therapy.

INTRODUCTION: Neurodegeneration is likely to be present from the earliest stages of multiple sclerosis (MS). MS responds poorly to disease-modifying treatments (DMTs) and leads to irreversible brain volume loss (BVL), which is a reliable predictor of future physical and cognitive disability. Our study aimed to discover the relationship between BVL, disease activity, and DMTs in a cohort of patients with MS. MATERIAL AND METHODS: A total of 147 patients fulfilled our inclusion criteria. Relevant demographic and clinical data (age, gender, time of MS onset, time of treatment initiation, DMT characteristics, Expanded Disability Status Scale (EDSS), number of relapses in the last two years prior to MRI examination) were correlated with MRI findings. RESULTS: Patients with progressive MS had significantly lower total brain and grey matter volumes (p = 0.003; p < 0.001), and higher EDSS scores (p < 0.001), compared to relapsing-remitting patients matched by disease duration and age. There was no association between MRI atrophy and MRI activity (c2 = 0.013, p = 0.910). Total EDSS negatively correlated with the whole brain (rs = -0.368, p < 0.001) and grey matter volumes (rs = -0.308, p < 0.001), but was not associated with the number of relapses in the last two years (p = 0.278). Delay in DMT negatively correlated with whole brain (rs = -0.387, p < 0.001) and grey matter volumes (rs = -0.377, p < 0.001). Treatment delay was connected with a higher risk for lower brain volume (b = -3.973, p < 0.001), and also predicted a higher EDSS score (b = 0.067, p < 0.001). CONCLUSIONS: Brain volume loss is a major contributor to disability progression, independent of disease activity. Delay in DMT leads to higher BVL and increased disability. Brain atrophy assessment should be translated into daily clinical practice to monitor disease course and response to DMTs. The assessment of BVL itself should be considered a suitable marker for treatment escalation.

Prospective classification of Alzheimer's disease conversion from mild cognitive impairment.

Alzheimer's disease (AD) is emerging as a serious problem with the rapid aging of the population, but due to the unclear cause of the disease and the absence of therapy, appropriate preventive measures are the next best thing. For this reason, it is important to early detect whether the disease converts from mild cognitive impairment (MCI) which is a prodromal phase of AD. With the advance in brain imaging techniques, various machine learning algorithms have become able to predict the conversion from MCI to AD by learning brain atrophy patterns. However, at the time of diagnosis, it is difficult to distinguish between the conversion group and the non-conversion group of subjects because the difference between groups is small, but the within-group variability is large in brain images. After a certain period of time, the subjects of conversion group show significant brain atrophy, whereas subjects of non-conversion group show only subtle changes due to the normal aging effect. This difference on brain atrophy makes the brain images more discriminative for learning. Motivated by this, we propose a method to perform classification by projecting brain images into the future, namely prospective classification. The experiments on the Alzheimer's Disease Neuroimaging Initiative dataset show that the prospective classification outperforms ordinary classification. Moreover, the features of prospective classification indicate the brain regions that significantly influence the conversion from MCI to AD.

Tau pathology as determinant of changes in atrophy and cerebral blood flow: a multi-modal longitudinal imaging study.

PURPOSE: Tau pathology is associated with concurrent atrophy and decreased cerebral blood flow (CBF) in Alzheimer's disease (AD), but less is known about their temporal relationships. Our aim was therefore to investigate the association of concurrent and longitudinal tau PET with longitudinal changes in atrophy and relative CBF. METHODS: We included 61 individuals from the Amsterdam Dementia Cohort (mean age 65.1 ± 7.5 years, 44% female, 57% amyloid-ß positive [Aß +], 26 cognitively impaired [CI]) who underwent dynamic [18F]flortaucipir PET and structural MRI at baseline and 25 ± 5 months follow-up. In addition, we included 86 individuals (68 CI) who only underwent baseline dynamic [18F]flortaucipir PET and MRI scans to increase power in our statistical models. We obtained [18F]flortaucipir PET binding potential (BPND) and R1 values reflecting tau load and relative CBF, respectively, and computed cortical thickness from the structural MRI scans using FreeSurfer. We assessed the regional associations between i) baseline and ii) annual change in tau PET BPND in Braak I, III/IV, and V/VI regions and cortical thickness or R1 in cortical gray matter regions (spanning the whole brain) over time using linear mixed models with random intercepts adjusted for age, sex, time between baseline and follow-up assessments, and baseline BPND in case of analyses with annual change as determinant. All analyses were performed in Aß- cognitively normal (CN) individuals and Aß+ (CN and CI) individuals separately. RESULTS: In Aß+ individuals, greater baseline Braak III/IV and V/VI tau PET binding was associated with faster cortical thinning in primarily frontotemporal regions. Annual changes in tau PET were not associated with cortical thinning over time in either Aß+ or Aß- individuals. Baseline tau PET was not associated with longitudinal changes in relative CBF, but increases in Braak III/IV tau PET over time were associated with increases in parietal relative CBF over time in Aß + individuals. CONCLUSION: We showed that higher tau load was related to accelerated cortical thinning, but not to decreases in relative CBF. Moreover, tau PET load at baseline was a stronger predictor of cortical thinning than change of tau PET signal.

MPS-FFA: A multiplane and multiscale feature fusion attention network for Alzheimer's disease prediction with structural MRI.

Structural magnetic resonance imaging (sMRI) is a popular technique that is widely applied in Alzheimer's disease (AD) diagnosis. However, only a few structural atrophy areas in sMRI scans are highly associated with AD. The degree of atrophy in patients' brain tissues and the distribution of lesion areas differ among patients. Therefore, a key challenge in sMRI-based AD diagnosis is identifying discriminating atrophy features. Hence, we propose a multiplane and multiscale feature-level fusion attention (MPS-FFA) model. The model has three components, (1) A feature encoder uses a multiscale feature extractor with hybrid attention layers to simultaneously capture and fuse multiple pathological features in the sagittal, coronal, and axial planes. (2) A global attention classifier combines clinical scores and two global attention layers to evaluate the feature impact scores and balance the relative contributions of different feature blocks. (3) A feature similarity discriminator minimizes the feature similarities among heterogeneous labels to enhance the ability of the network to discriminate atrophy features. The MPS-FFA model provides improved interpretability for identifying discriminating features using feature visualization. The experimental results on the baseline sMRI scans from two databases confirm the effectiveness (e.g., accuracy and generalizability) of our method in locating pathological locations. The source code is available at https://github.com/LiuFei-AHU/MPSFFA.

Multiclass prediction of different dementia syndromes based on multi-centric volumetric MRI imaging.

INTRODUCTION: Dementia syndromes can be difficult to diagnose. We aimed at building a classifier for multiple dementia syndromes using magnetic resonance imaging (MRI). METHODS: Atlas-based volumetry was performed on T1-weighted MRI data of 426 patients and 51 controls from the multi-centric German Research Consortium of Frontotemporal Lobar Degeneration including patients with behavioral variant frontotemporal dementia, Alzheimer's disease, the three subtypes of primary progressive aphasia, i.e., semantic, logopenic and nonfluent-agrammatic variant, and the atypical parkinsonian syndromes progressive supranuclear palsy and corticobasal syndrome. Support vector machine classification was used to classify each patient group against controls (binary classification) and all seven diagnostic groups against each other in a multi-syndrome classifier (multiclass classification). RESULTS: The binary classification models reached high prediction accuracies between 71 and 95% with a chance level of 50%. Feature importance reflected disease-specific atrophy patterns. The multi-syndrome model reached accuracies of more than three times higher than chance level but was far from 100%. Multi-syndrome model performance was not homogenous across dementia syndromes, with better performance in syndromes characterized by regionally specific atrophy patterns. Whereas diseases generally could be classified vs controls more correctly with increasing severity and duration, differentiation between diseases was optimal in disease-specific windows of severity and duration. DISCUSSION: Results suggest that automated methods applied to MR imaging data can support physicians in diagnosis of dementia syndromes. It is particularly relevant for orphan diseases beside frequent syndromes such as Alzheimer's disease.

Generative adversarial network constrained multiple loss autoencoder: A deep learning-based individual atrophy detection for Alzheimer's disease and mild cognitive impairment.

Exploring individual brain atrophy patterns is of great value in precision medicine for Alzheimer's disease (AD) and mild cognitive impairment (MCI). However, the current individual brain atrophy detection models are deficient. Here, we proposed a framework called generative adversarial network constrained multiple loss autoencoder (GANCMLAE) for precisely depicting individual atrophy patterns. The GANCMLAE model was trained using normal controls (NCs) from the Alzheimer's Disease Neuroimaging Initiative cohort, and the Xuanwu cohort was employed to validate the robustness of the model. The potential of the model for identifying different atrophy patterns of MCI subtypes was also assessed. Furthermore, the clinical application potential of the GANCMLAE model was investigated. The results showed that the model can achieve good image reconstruction performance on the structural similarity index measure (0.929 ± 0.003), peak signal-to-noise ratio (31.04 ± 0.09), and mean squared error (0.0014 ± 0.0001) with less latent loss in the Xuanwu cohort. The individual atrophy patterns extracted from this model are more precise in reflecting the clinical symptoms of MCI subtypes. The individual atrophy patterns exhibit a better discriminative power in identifying patients with AD and MCI from NCs than those of the t-test model, with areas under the receiver operating characteristic curve of 0.867 (95%: 0.837-0.897) and 0.752 (95%: 0.71-0.790), respectively. Similar findings are also reported in the AD and MCI subgroups. In conclusion, the GANCMLAE model can serve as an effective tool for individualised atrophy detection.

Structural alterations of spinocerebellar ataxias type 3: from pre-symptomatic to symptomatic stage.

OBJECTIVES: To investigate and characterize the structural alterations of the brain in SCA3, and their correlations with the scale for the assessment and rating of ataxia (SARA) and normal brain ATXN3 expression. METHODS: We performed multimodal analyses in 52 SCA3 (15 pre-symptomatic) and healthy controls (HCs) (n = 35) to assess the abnormalities of gray and white matter (WM) of the cerebrum, brainstem, and cerebellum via FreeSurfer, SUIT, and TBSS, and their associations with disease severity. Twenty SCA3 patients (5 pre- and 15 symptomatic) were followed for at least a year. Besides, we uncovered the normal pattern of brain ATXN3 spatial distribution. RESULTS: Pre-symptomatic patients showed only WM damage, mainly in the cerebellar peduncles, compared to HCs. In the advanced stage, the WM damage followed a caudal-rostral pattern. Meanwhile, continuous nonlinear structure damage was characterized by brainstem volumetric reduction and relatively symmetric cerebellar and basal ganglia atrophy but spared the cerebral cortex, partially explained by the ATXN3 overexpression. The bilateral pallidum, brainstem, and cerebellar peduncles demonstrated a very large effect size. Besides, all these alterations were significantly correlated with SARA; the pons (r = -0.65) and superior cerebellar peduncle (r = -0.68) volume demonstrated a higher correlation than the cerebellum with SARA. The longitudinal study further uncovered progressive atrophy of pons in symptomatic SCA3. CONCLUSIONS: Significant WM damage starts before the ataxia onset. The bilateral pallidum, brainstem, and cerebellar peduncles are the most vulnerable targets. The volume of pons appears to be the most promising imaging biomarker for a longitudinal study. TRIAL REGISTRATION: ClinicalTrial ID: ChiCTR2100045857 ( http://www.chictr.org.cn/edit.aspx?pid=55652&htm=4 ) KEY POINTS: • Pre- SCA3 showed WM damage mainly in cerebellar peduncles. Continuous brain damage was characterized by brainstem, widespread, and relatively symmetric cerebellar and basal ganglia atrophy. • Volumetric abnormalities were most evident in the bilateral pallidum, brainstem, and cerebellar peduncles in SCA3. • The volume of pons might identify the disease progression longitudinally.

Selective atrophy of the cervical enlargement in whole spinal cord MRI of amyotrophic lateral sclerosis.

Amyotrophic lateral sclerosis (ALS) is a deadly neurodegenerative disorder affecting motor neurons in the spinal cord and brain. Studies have reported on atrophy within segments of the cervical cord, but we are not aware of previous investigations of the whole spinal cord. Herein we present our findings from a 3T MRI study involving 32 subjects (15 ALS participants and 17 healthy controls) characterizing cross-sectional area along the entire cord. We report atrophy of the cervical enlargement in ALS participants, but no evidence of atrophy of the thoracolumbar enlargement. These results suggest that MR-based analyses of the cervical cord may be sufficient for in vivo investigations of spinal cord atrophy in ALS, and that atrophy of the cervical enlargement (C4-C7) is a potential imaging marker for quantifying lower motor neuron degradation.

Related amyloid burden and cortical atrophy in individuals with subtle cognitive decline.

BACKGROUND AND PURPOSE: Subtle cognitive decline represents a stage of cognitive deterioration in which pathological biomarkers may be present, including early cortical atrophy and amyloid deposition. Using individual items from the Montreal Cognitive Assessment and k-modes cluster analysis, we previously identified three clusters of individuals without overt cognitive impairment: (1) High Performing (no deficits in performance), (2) Memory Deficits (lower memory performance), and (3) Compound Deficits (lower memory and executive function performance). In this study, we sought to understand the relationships found in our clusters between cortical atrophy on MR and amyloid burden on PET. METHODS: Data were derived from the Alzheimer's Disease Neuroimaging Initiative and comprised individuals from our previous analyses with available MR and amyloid PET scans (n = 272). Using multiple-group structural equation modeling, we regressed amyloid standardized uptake value ratio on volumetric regions to simultaneously evaluate unique associations within each cluster. RESULTS: In our Compound Deficits cluster, greater whole cerebral amyloid burden was significantly related to right entorhinal cortical and left hippocampal atrophy, rs  = -.412 (p = .005) and -.304 (p = .049), respectively. Within this cluster, right entorhinal cortical atrophy was significantly related to greater amyloid burden within multiple frontal regions. CONCLUSIONS: The Compound Deficits cluster, which represents a group potentially at higher risk for decline, was observed to have significantly more cortical atrophy, particularly within the entorhinal cortex and hippocampus, associated with whole brain and frontal lobe amyloid burden. These findings point to a pattern of early pathological deterioration that may place these individuals at risk for future decline.

Clinical and metabolic imaging features of late-onset and early-onset posterior cortical atrophy.

BACKGROUND AND PURPOSE: Late-onset (LO) and early-onset (EO) dementia show neurobiological and clinical differences. Clinical and 18 fluoro-deoxy-glucose positron emission tomography (FDG-PET) features of LO and EO posterior cortical atrophy (LO_PCA, EO_PCA), the visual variant of Alzheimer's disease (AD), were compared. LO_PCA patients were also compared with a group of patients with LO typical AD (tAD). METHODS: Thirty-seven LO_PCA patients (onset age ≥ 65 years), 29 EO_PCA patients and 40 tAD patients who all underwent a standard neuropsychological battery were recruited; PCA patients were also assessed for the presence of posterior signs and symptoms. Brain FDG-PET was available in 32 LO_PCA cases, 23 EO_PCA cases and all tAD cases, and their scans were compared with scans from 30 healthy elderly controls. Within the entire PCA sample FDG uptake was also correlated with age at onset as a continuous variable. RESULTS: The main difference between the two PCA groups was a higher prevalence of Bálint-Holmes symptoms in EO cases, which was associated with the presence of severe bilateral occipito-temporo-parietal hypometabolism, whilst LO_PCA patients showed reduction of FDG uptake mainly in the right posterior regions. The latter group also showed mesial temporal hypometabolism, similarly to the tAD group, although with a right rather than left lateralization. Correlation analysis confirmed the association between older age and decreased limbic metabolism and between younger age and decreased left parietal metabolism. CONCLUSIONS: The major involvement of the temporal cortex in LO cases and of the parietal cortex in EO cases reported previously within the AD spectrum holds true also for the visual variant of AD.

Machine learning-based automatic estimation of cortical atrophy using brain computed tomography images.

Cortical atrophy is measured clinically according to established visual rating scales based on magnetic resonance imaging (MRI). Although brain MRI is the primary imaging marker for neurodegeneration, computed tomography (CT) is also widely used for the early detection and diagnosis of dementia. However, they are seldom investigated. Therefore, we developed a machine learning algorithm for the automatic estimation of cortical atrophy on brain CT. Brain CT images (259 Alzheimer's dementia and 55 cognitively normal subjects) were visually rated by three neurologists and used for training. We constructed an algorithm by combining the convolutional neural network and regularized logistic regression (RLR). Model performance was then compared with that of neurologists, and feature importance was measured. RLR provided fast and reliable automatic estimations of frontal atrophy (75.2% accuracy, 93.6% sensitivity, 67.2% specificity, and 0.87 area under the curve [AUC]), posterior atrophy (79.6% accuracy, 87.2% sensitivity, 75.9% specificity, and 0.88 AUC), right medial temporal atrophy (81.2% accuracy, 84.7% sensitivity, 79.6% specificity, and 0.88 AUC), and left medial temporal atrophy (77.7% accuracy, 91.1% sensitivity, 72.3% specificity, and 0.90 AUC). We concluded that RLR-based automatic estimation of brain CT provided a comprehensive rating of atrophy that can potentially support physicians in real clinical settings.

Automated brain volumetric measures with AccuBrain: version comparison in accuracy, reproducibility and application for diagnosis.

BACKGROUND: Automated brain volumetry has been widely used to assess brain volumetric changes that may indicate clinical states and progression. Among the tools that implement automated brain volumetry, AccuBrain has been validated for its accuracy, reliability and clinical applications for the older version (IV1.2). Here, we aim to investigate the performance of an updated version (IV2.0) of AccuBrain for future use from several aspects. METHODS: Public datasets with 3D T1-weighted scans were included for version comparisons, each with Alzheimer's disease (AD) patients and normal control (NC) subjects that were matched in age and gender. For the comparisons of the brain volumetric measures quantified from the same scans, we investigated the difference of hippocampal segmentation accuracy (using Dice similarity coefficient [DSC] as the major measurement). As AccuBrain generates a composite index (AD resemblance atrophy index, AD-RAI) that indicates similarity with AD-like brain atrophy pattern, we also compared the two versions for the diagnostic accuracy of AD versus NC with AD-RAI. Also, we examined the intra-scanner reproducibility of the two versions for the scans acquired with short-intervals using intraclass correlation coefficient. RESULTS: AccuBrain IV2.0 presented significantly higher accuracy of hippocampal segmentation (DSC: 0.91 vs. 0.89, p < 0.001) and diagnostic accuracy of AD (AUC: 0.977 vs. 0.921, p < 0.001) than IV1.2. The results of intra-scanner reproducibility did not favor one version over the other. CONCLUSIONS: AccuBrain IV2.0 presented better segmentation accuracy and diagnostic accuracy of AD, and similar intra-scanner reproducibility compared with IV1.2. Both versions should be feasible for use due to the small magnitude of differences.