Cancer cell-specific and pro-apoptotic SMAC peptide-doxorubicin conjugated prodrug encapsulated aposomes for synergistic cancer immunotherapy.

BACKGROUND: Immunogenic cell death (ICD) is a crucial approach to turn immunosuppressive tumor microenvironment (ITM) into immune-responsive milieu and improve the response rate of immune checkpoint blockade (ICB) therapy. However, cancer cells show resistance to ICD-inducing chemotherapeutic drugs, and non-specific toxicity of those drugs against immune cells reduce the immunotherapy efficiency. METHODS: Herein, we propose cancer cell-specific and pro-apoptotic liposomes (Aposomes) encapsulating second mitochondria-derived activator of caspases mimetic peptide (SMAC-P)-doxorubicin (DOX) conjugated prodrug to potentiate combinational ICB therapy with ICD. The SMAC-P (AVPIAQ) with cathepsin B-cleavable peptide (FRRG) was directly conjugated to DOX, and the resulting SMAC-P-FRRG-DOX prodrug was encapsulated into PEGylated liposomes. RESULTS: The SMAC-P-FRRG-DOX encapsulated PEGylated liposomes (Aposomes) form a stable nanostructure with an average diameter of 109.1 ± 5.14 nm and promote the apoptotic cell death mainly in cathepsin B-overexpressed cancer cells. Therefore, Aposomes induce a potent ICD in targeted cancer cells in synergy of SMAC-P with DOX in cultured cells. In colon tumor models, Aposomes efficiently accumulate in targeted tumor tissues via enhanced permeability and retention (EPR) effect and release the encapsulated prodrug of SMAC-P-FRRG-DOX, which is subsequently cleaved to SMAC-P and DOX in cancer cells. Importantly, the synergistic activity of inhibitors of apoptosis proteins (IAPs)-inhibitory SMAC-P sensitizing the effects of DOX induces a potent ICD in the cancer cells to promote dendritic cell (DC) maturation and stimulate T cell proliferation and activation, turning ITM into immune-responsive milieu. CONCLUSIONS: Eventually, the combination of Aposomes with anti-PD-L1 antibody results in a high rate of complete tumor regression (CR: 80%) and also prevent the tumor recurrence by immunological memory established during treatments.

Effect of Amyloid on Cognitive Performance in Parkinson's Disease and Dementia with Lewy Bodies.

BACKGROUND: Concomitant amyloid pathology contributes to the clinical heterogeneity of Lewy body diseases (LBDs). OBJECTIVE: The objective of this study was to investigate the pattern and effect of amyloid accumulation on cognitive dysfunction in Parkinson's disease (PD) and dementia with Lewy bodies (DLB). METHODS: We retrospectively assessed 205 patients with LBD (91 with DLB and 114 with PD) who underwent 18 F-florbetaben positron emission tomography and divided them into amyloid-positive and amyloid-negative groups depending on global standardized uptake value ratios (SUVRs). We investigated the effect of group on the regional and global SUVRs using general linear models (GLMs) after controlling for age, sex, cognitive status, and score on the Korean version of the Mini-Mental State Examination. Moreover, the effect of amyloid on cognitive function, depending on the type of LBD, was evaluated using GLMs with interaction analysis. RESULTS: In all evaluated regions including the striatum, the DLB group showed a higher SUVR than the PD group. Among amyloid-positive patients, the DLB group had a higher regional SUVR than the PD group in the frontal and parietal cortices. There was a significant interaction effect between amyloid and disease groups in language and memory function. In patients with PD, global amyloid load was negatively associated with language (B = -2.03; P = 0.010) and memory functions (B = -1.96; P < 0.001). However, amyloid load was not significantly associated with cognitive performance in the DLB group. CONCLUSIONS: Although the burden of amyloid was higher in the DLB group, amyloid accumulation was negatively associated with the memory and language functions in the PD group only. © 2022 International Parkinson and Movement Disorder Society.

Harmonisation of PET imaging features with different amyloid ligands using machine learning-based classifier.

PURPOSE: In this study, we used machine learning to develop a new method derived from a ligand-independent amyloid (Aß) positron emission tomography (PET) classifier to harmonise different Aß ligands. METHODS: We obtained 107 paired 18F-florbetaben (FBB) and 18F-flutemetamol (FMM) PET images at the Samsung Medical Centre. To apply the method to FMM ligand, we transferred the previously developed FBB PET classifier to test similar features from the FMM PET images for application to FMM, which in turn developed a ligand-independent Aß PET classifier. We explored the concordance rates of our classifier in detecting cortical and striatal Aß positivity. We investigated the correlation of machine learning-based cortical tracer uptake (ML-CTU) values quantified by the classifier between FBB and FMM. RESULTS: This classifier achieved high classification accuracy (area under the curve = 0.958) even with different Aß PET ligands. In addition, the concordance rate of FBB and FMM using the classifier (87.5%) was good to excellent, which seemed to be higher than that in visual assessment (82.7%) and lower than that in standardised uptake value ratio cut-off categorisation (93.3%). FBB and FMM ML-CTU values were highly correlated with each other (R = 0.903). CONCLUSION: Our findings suggested that our novel classifier may harmonise FBB and FMM ligands in the clinical setting which in turn facilitate the biomarker-guided diagnosis and trials of anti-Aß treatment in the research field.

Local shape volume alterations in subcortical structures of suicide attempters with major depressive disorder.

Suicide is among the most important global health concerns; accordingly, an increasing number of studies have shown the risks for suicide attempt(s) in terms of brain morphometric features and their clinical correlates. However, brain studies addressing suicidal vulnerability have been more focused on demonstrating impairments in cortical structures than in the subcortical structures. Using local shape volumes (LSV) analysis, we investigated subcortical structures with their clinical correlates in depressed patients who attempted suicide. Then we compared them with depressed patients without a suicidal history and age- and sex-matched healthy controls (HCs; i.e., 47 suicide attempters with depression, 47 non-suicide attempters with depression, and 109 HCs). Significant volumetric differences were found between suicidal and nonsuicidal depressed patients in several vertices: 16 in the left amygdala; 201 in the left hippocampus; 1,057 in the left putamen; and 140 in the left pallidum; 1 in the right pallidum; and 6 in the bilateral thalamus. These findings indicated subcortical alterations in LSV in components of the limbic-cortical-striatal-pallidal-thalamic circuits. Moreover, our results demonstrated that the basal ganglia was correlated with perceived stress levels, and the thalamus was correlated with suicidal ideation. We suggest that suicidality in major depressive disorder may involve subcortical volume alterations.

Correction to: Staging and quantification of florbetaben PET images using machine learning: impact of predicted regional cortical tracer uptake and amyloid stage on clinical outcomes.

Incorrect acknowledgments.

Staging and quantification of florbetaben PET images using machine learning: impact of predicted regional cortical tracer uptake and amyloid stage on clinical outcomes.

PURPOSE: We developed a machine learning-based classifier for in vivo amyloid positron emission tomography (PET) staging, quantified cortical uptake of the PET tracer by using a machine learning method, and investigated the impact of these amyloid PET parameters on clinical and structural outcomes. METHODS: A total of 337 18F-florbetaben PET scans obtained at Samsung Medical Center were assessed. We defined a feature vector representing the change in PET tracer uptake from grey to white matter. Using support vector machine (SVM) regression and SVM classification, we quantified the cortical uptake as predicted regional cortical tracer uptake (pRCTU) and categorised the scans as positive and negative. Positive scans were further classified into two stages according to the striatal uptake. We compared outcome parameters among stages and further assessed the association between the pRCTU and outcome variables. Finally, we performed path analysis to determine mediation effects between PET variables. RESULTS: The classification accuracy was 97.3% for cortical amyloid positivity and 91.1% for striatal positivity. The left frontal and precuneus/posterior cingulate regions, as well as the anterior portion of the striatum, were important in determination of stages. The clinical scores and magnetic resonance imaging parameters showed negative associations with PET stage. However, except for the hippocampal volume, most outcomes were associated with the stage through the complete mediation effect of pRCTU. CONCLUSION: Using a machine learning algorithm, we achieved high accuracy for in vivo amyloid PET staging. The in vivo amyloid stage was associated with cognitive function and cerebral atrophy mostly through the mediation effect of cortical amyloid.

Correction to: Staging and quantification of florbetaben PET images using machine learning: impact of predicted regional cortical tracer uptake and amyloid stage on clinical outcomes.

The Table 2 in the original version of this article contained a mistake in the alignment. Correct Table 2 presentation is presented here.

Application of an amyloid and tau classification system in subcortical vascular cognitive impairment patients.

OBJECTIVE: To apply an AT (Aß/tau) classification system to subcortical vascular cognitive impairment (SVCI) patients following recently developed biomarker-based criteria of Alzheimer's disease (AD), and to investigate its clinical significance. METHODS: We recruited 60 SVCI patients who underwent the neuropsychological tests, brain MRI, and 18F-florbetaben and 18F-AV1451 PET at baseline. As a control group, we further recruited 27 patients with AD cognitive impairment (ADCI; eight Aß PET-positive AD dementia and 19 amnestic mild cognitive impairment). ADCI and SVCI patients were classified as having normal or abnormal Aß (A-/A+) and tau (T-/T+) based on PET results. Across the three SVCI groups (A-, A+T-, and A+T+SVCI), we compared longitudinal changes in cognition, hippocampal volume (HV), and cortical thickness using linear mixed models. RESULTS: Among SVCI patients, 33 (55%), 20 (33.3%), and seven (11.7%) patients were A-, A+T-, and A+T+, respectively. The frequency of T+ was lower in A+SVCI (7/27, 25.9%) than in A+ADCI (14/20, 70.0%, p = 0.003) which suggested that cerebral small vessel disease affected cognitive impairments independently of A+. A+T-SVCI had steeper cognitive decline than A-SVCI. A+T+SVCI also showed steeper cognitive decline than A+T-SVCI. Also, A+T-SVCI had steeper decrease in HV than A-SVCI, while cortical thinning did not differ between the two groups. A+T+SVCI had greater global cortical thinning compared with A+T-SVCI, while declines in HV did not differ between the two groups. CONCLUSION: This study showed that the AT system successfully characterized SVCI patients, suggesting that the AT system may be usefully applied in a research framework for clinically diagnosed SVCI.

Evaluation of Diffusion Lesion Volume Measurements in Acute Ischemic Stroke Using Encoder-Decoder Convolutional Network.

Background and Purpose- Automatic segmentation of cerebral infarction on diffusion-weighted imaging (DWI) is typically performed based on a fixed apparent diffusion coefficient (ADC) threshold. Fixed ADC threshold methods may not be accurate because ADC values vary over time after stroke onset. Deep learning has the potential to improve the accuracy, provided that a large set of correctly annotated lesion data is used for training. The purpose of this study was to evaluate deep learning-based methods and compare them with commercial software in terms of lesion volume measurements. Methods- U-net, an encoder-decoder convolutional neural network, was adopted to train segmentation models. Two U-net models were developed: a U-net (DWI+ADC) model, trained on DWI and ADC data, and a U-net (DWI) model, trained on DWI data only. A total of 296 subjects were used for training and 134 for external validation. An expert neurologist manually delineated the stroke lesions on DWI images, which were used as the ground-truth reference. Lesion volume measurements from the U-net methods were compared against the expert's manual segmentation and Rapid Processing of Perfusion and Diffusion (RAPID; iSchemaView Inc) analysis. Results- In external validation, U-net (DWI+ADC) showed the highest intraclass correlation coefficient with manual segmentation (intraclass correlation coefficient, 1.0; 95% CI, 0.99-1.00) and sufficiently high correlation with the RAPID results (intraclass correlation coefficient, 0.99; 95% CI, 0.98-0.99). U-net (DWI+ADC) and manual segmentation resulted in the smallest 95% Bland-Altman limits of agreement (-5.31 to 4.93 mL) with a mean difference of -0.19 mL. Conclusions- The presented deep learning-based method is fully automatic and shows a high correlation of diffusion lesion volume measurements with manual segmentation and commercial software. The method has the potential to be used in patient selection for endovascular reperfusion therapy in the late time window of acute stroke.

Amyloid involvement in subcortical regions predicts cognitive decline.

PURPOSE: We estimated whether amyloid involvement in subcortical regions may predict cognitive impairment, and established an amyloid staging scheme based on degree of subcortical amyloid involvement. METHODS: Data from 240 cognitively normal older individuals, 393 participants with mild cognitive impairment, and 126 participants with Alzheimer disease were acquired at Alzheimer's Disease Neuroimaging Initiative sites. To assess subcortical involvement, we analyzed amyloid deposition in amygdala, putamen, and caudate nucleus. We staged participants into a 3-stage model based on cortical and subcortical amyloid involvement: 382 with no cortical or subcortical involvement as stage 0, 165 with cortical but no subcortical involvement as stage 1, and 203 with both cortical and subcortical involvement as stage 2. RESULTS: Amyloid accumulation was first observed in cortical regions and spread down to the putamen, caudate nucleus, and amygdala. In longitudinal analysis, changes in MMSE, ADAS-cog 13, FDG PET SUVR, and hippocampal volumes were steepest in stage 2 followed by stage 1 then stage 0 (p value <0.001). Stage 2 showed steeper changes in MMSE score (ß [SE] = -0.02 [0.004], p < 0.001), ADAS-cog 13 (0.05 [0.01], p < 0.001), FDG PET SUVR (-0.0008 [0.0003], p = 0.004), and hippocampal volumes (-4.46 [0.65], p < 0.001) compared to stage 1. CONCLUSIONS: We demonstrated a downward spreading pattern of amyloid, suggesting that amyloid accumulates first in neocortex followed by subcortical structures. Furthermore, our new finding suggested that an amyloid staging scheme based on subcortical involvement might reveal how differential regional accumulation of amyloid affects cognitive decline through functional and structural changes of the brain.

Determination of Optimal Heart Rate Variability Features Based on SVM-Recursive Feature Elimination for Cumulative Stress Monitoring Using ECG Sensor.

Routine stress monitoring in daily life can predict potentially serious health impacts. Effective stress monitoring in medical and healthcare fields is dependent upon accurate determination of stress-related features. In this study, we determined the optimal stress-related features for effective monitoring of cumulative stress. We first investigated the effects of short- and long-term stress on various heart rate variability (HRV) features using a rodent model. Subsequently, we determined an optimal HRV feature set using support vector machine-recursive feature elimination (SVM-RFE). Experimental results indicate that the HRV time domain features generally decrease under long-term stress, and the HRV frequency domain features have substantially significant differences under short-term stress. Further, an SVM classifier with a radial basis function kernel proved most accurate (93.11%) when using an optimal HRV feature set comprising the mean of R-R intervals (mRR), the standard deviation of R-R intervals (SDRR), and the coefficient of variance of R-R intervals (CVRR) as time domain features, and the normalized low frequency (nLF) and the normalized high frequency (nHF) as frequency domain features. Our findings indicate that the optimal HRV features identified in this study can effectively and efficiently detect stress. This knowledge facilitates development of in-facility and mobile healthcare system designs to support stress monitoring in daily life.

Normalization of cortical thickness measurements across different T1 magnetic resonance imaging protocols by novel W-Score standardization.

BACKGROUND: The use of different 3D T1-weighted magnetic resonance (T1 MR) imaging protocols induces image incompatibility across multicenter studies, negating the many advantages of multicenter studies. A few methods have been developed to address this problem, but significant image incompatibility still remains. Thus, we developed a novel and convenient method to improve image compatibility. METHODS: W-score standardization creates quality reference values by using a healthy group to obtain normalized disease values. We developed a protocol-specific w-score standardization to control the protocol effect, which is applied to each protocol separately. We used three data sets. In dataset 1, brain T1 MR images of normal controls (NC) and patients with Alzheimer's disease (AD) from two centers, acquired with different T1 MR protocols, were used (Protocol 1 and 2, n = 45/group). In dataset 2, data from six subjects, who underwent MRI with two different protocols (Protocol 1 and 2), were used with different repetition times, echo times, and slice thicknesses. In dataset 3, T1 MR images from a large number of healthy normal controls (Protocol 1: n = 148, Protocol 2: n = 343) were collected for w-score standardization. The protocol effect and disease effect on subjects' cortical thickness were analyzed before and after the application of protocol-specific w-score standardization. RESULTS: As expected, different protocols resulted in differing cortical thickness measurements in both NC and AD subjects. Different measurements were obtained for the same subject when imaged with different protocols. Multivariate pattern difference between measurements was observed between the protocols. Classification accuracy between two protocols was nearly 90%. After applying protocol-specific w-score standardization, the differences between the protocols substantially decreased. Most importantly, protocol-specific w-score standardization reduced both univariate and multivariate differences in the images while maintaining the AD disease effect. Compared to conventional regression methods, our method showed the best performance for in terms of controlling the protocol effect while preserving disease information. CONCLUSIONS: Protocol-specific w-score standardization effectively resolved the concerns of conventional regression methods. It showed the best performance for improving the compatibility of a T1 MR post-processed feature, cortical thickness.

Subcortical shape analysis of progressive mild cognitive impairment in Parkinson's disease.

BACKGROUND: Cortical neural correlates of ongoing cognitive decline in Parkinson's disease (PD) have been suggested; however, the role of subcortical structures in longitudinal change of cognitive dysfunction in PD has not been fully investigated. Here, we used automatic analysis to explore subcortical brain structures in patients with PD with mild cognitive impairment that converts into PD with dementia. METHODS: One hundred eighty-two patients with PD with mild cognitive impairment were classified as PD with mild cognitive impairment converters (n = 74) or nonconverters (n = 108), depending on whether they were subsequently diagnosed with dementia in PD. We used surface-based analysis to compare atrophic changes of subcortical brain structures between PD with mild cognitive impairment converters and nonconverters. RESULTS: PD with mild cognitive impairment converters had lower cognitive composite scores in the attention and frontal executive domains than did nonconverters. Subcortical shape analysis revealed that PD with mild cognitive impairment converters had smaller local shape volumes than did nonconverters in the bilateral thalamus, right caudate, and right hippocampus. Logistic regression analysis showed that local shape volumes in the bilateral thalamus and right caudate were significant independent predictors of PD with mild cognitive impairment converters. In the PD with mild cognitive impairment converter group, thalamic local shape volume was associated with semantic fluency and attentional composite score. CONCLUSIONS: The present data suggest that the local shape volumes of deep subcortical structures, especially in the caudate and thalamus, may serve as important predictors of the development of dementia in patients with PD. © 2017 International Parkinson and Movement Disorder Society.

Amyloid burden, cerebrovascular disease, brain atrophy, and cognition in cognitively impaired patients.

BACKGROUND: We investigated the independent effects of Alzheimer's disease (AD) and cerebrovascular disease (CVD) pathologies on brain structural changes and cognition. METHODS: Amyloid burden (Pittsburgh compound B [PiB] retention ratio), CVD markers (volume of white matter hyperintensities [WMH] and number of lacunae), and structural changes (cortical thickness and hippocampal shape) were measured in 251 cognitively impaired patients. Path analyses were utilized to assess the effects of these markers on cognition. RESULTS: PiB retention ratio was associated with hippocampal atrophy, which was associated with memory impairment. WMH were associated with frontal thinning, which was associated with executive and memory dysfunctions. PiB retention ratio and lacunae were also associated with memory and executive dysfunction without the mediation of hippocampal or frontal atrophy. CONCLUSIONS: Our results suggest that the impacts of AD and CVD pathologies on cognition are mediated by specific brain regions.

Deep learning-based dynamic ventilatory threshold estimation from electrocardiograms.

BACKGROUND AND OBJECTIVE: The ventilatory threshold (VT) marks the transition from aerobic to anaerobic metabolism and is used to assess cardiorespiratory endurance. A conventional way to assess VT is cardiopulmonary exercise testing, which requires a gas analyzer. Another method for measuring VT involves calculating the heart rate variability (HRV) from an electrocardiogram (ECG) by computing the variability of heartbeats. However, the HRV method has some limitations. ECGs should be recorded for at least 5 minutes to calculate the HRV, and the result may depend on the utilized ECG preprocessing algorithms. METHODS: To overcome these problems, we developed a deep learning-based model consisting of long short-term memory (LSTM) and convolutional neural network (CNN) for a lead II ECG. Variables reflecting subjects' physical characteristics, as well as ECG signals, were input into the model to estimate VT. We applied joint optimization to the CNN layers to generate an informative latent space, which was fed to the LSTM layers. The model was trained and evaluated on two datasets, one from the Bruce protocol and the other from a protocol including multiple tasks (MT). RESULTS: Acceptable performances (mean and 95% CI) were obtained on the datasets from the Bruce protocol (-0.28[-1.91,1.34] ml/min/kg) and the MT protocol (0.07[-3.14,3.28] ml/min/kg) regarding the differences between the predictions and labels. The coefficient of determination, Pearson correlation coefficient, and root mean square error were 0.84, 0.93, and 0.868 for the Bruce protocol and 0.73, 0.97, and 3.373 for the MT protocol, respectively. CONCLUSIONS: The results indicated that it is possible for the proposed model to simultaneously assess VT with the inputs of successive ECGs. In addition, from ablation studies concerning the physical variables and the joint optimization process, it was demonstrated that their use could boost the VT assessment performance of the model. The proposed model enables dynamic VT estimation with ECGs, which could help with managing cardiorespiratory fitness in daily life and cardiovascular rehabilitation in patients.

Differentiation between multiple sclerosis and neuromyelitis optica spectrum disorder using a deep learning model.

Multiple sclerosis (MS) and neuromyelitis optica spectrum disorder (NMOSD) are autoimmune inflammatory disorders of the central nervous system (CNS) with similar characteristics. The differential diagnosis between MS and NMOSD is critical for initiating early effective therapy. In this study, we developed a deep learning model to differentiate between multiple sclerosis (MS) and neuromyelitis optica spectrum disorder (NMOSD) using brain magnetic resonance imaging (MRI) data. The model was based on a modified ResNet18 convolution neural network trained with 5-channel images created by selecting five 2D slices of 3D FLAIR images. The accuracy of the model was 76.1%, with a sensitivity of 77.3% and a specificity of 74.8%. Positive and negative predictive values were 76.9% and 78.6%, respectively, with an area under the curve of 0.85. Application of Grad-CAM to the model revealed that white matter lesions were the major classifier. This compact model may aid in the differential diagnosis of MS and NMOSD in clinical practice.

Real-world prediction of preclinical Alzheimer's disease with a deep generative model.

Amyloid positivity is an early indicator of Alzheimer's disease and is necessary to determine the disease. In this study, a deep generative model is utilized to predict the amyloid positivity of cognitively normal individuals using proxy measures, such as structural MRI scans, demographic variables, and cognitive scores, instead of invasive direct measurements. Through its remarkable efficacy in handling imperfect datasets caused by missing data or labels, and imbalanced classes, the model outperforms previous studies and widely used machine learning approaches with an AUROC of 0.8609. Furthermore, this study illuminates the model's adaptability to diverse clinical scenarios, even when feature sets or diagnostic criteria differ from the training data. We identify the brain regions and variables that contribute most to classification, including the lateral occipital lobes, posterior temporal lobe, and APOE ϵ4 allele. Taking advantage of deep generative models, our approach can not only provide inexpensive, non-invasive, and accurate diagnostics for preclinical Alzheimer's disease, but also meet real-world requirements for clinical translation of a deep learning model, including transferability and interpretability.

Sex differences in the structural rich-club connectivity in patients with Alzheimer's disease.

Background and objectives: Alzheimer's disease (AD) is more prevalent in women than in men; however, there is a discrepancy in research on sex differences in AD. The human brain is a large-scale network with hub regions forming a central core, the rich-club, which is vital to cognitive functions. However, it is unknown whether alterations in the rich-clubs in AD differ between men and women. We aimed to investigate sex differences in the rich-club organization in the brains of patients with AD. Methods: In total, 260 cognitively unimpaired individuals with negative amyloid positron emission tomography (PET) scans, 281 with prodromal AD (mild cognitive impairment due to AD) and 285 with AD dementia who confirmed with positive amyloid PET scans participated in the study. We obtained high-resolution T1-weighted and diffusion tensor images and performed network analysis. Results: We observed sex differences in the rich-club and feeder connections in patients with AD, suggesting lower structural connectivity strength in women than in men. We observed a significant group-by-sex interaction in the feeder connections, particularly in the thalamus. In addition, the connectivity strength of the thalamus in the feeder connections was significantly correlated with general cognitive function in only men with prodromal AD and women with AD dementia. Conclusion: Our findings provide important evidence for sex-specific alterations in the structural brain network related to AD.

Individual subject classification for Alzheimer's disease based on incremental learning using a spatial frequency representation of cortical thickness data.

Patterns of brain atrophy measured by magnetic resonance structural imaging have been utilized as significant biomarkers for diagnosis of Alzheimer's disease (AD). However, brain atrophy is variable across patients and is non-specific for AD in general. Thus, automatic methods for AD classification require a large number of structural data due to complex and variable patterns of brain atrophy. In this paper, we propose an incremental method for AD classification using cortical thickness data. We represent the cortical thickness data of a subject in terms of their spatial frequency components, employing the manifold harmonic transform. The basis functions for this transform are obtained from the eigenfunctions of the Laplace-Beltrami operator, which are dependent only on the geometry of a cortical surface but not on the cortical thickness defined on it. This facilitates individual subject classification based on incremental learning. In general, methods based on region-wise features poorly reflect the detailed spatial variation of cortical thickness, and those based on vertex-wise features are sensitive to noise. Adopting a vertex-wise cortical thickness representation, our method can still achieve robustness to noise by filtering out high frequency components of the cortical thickness data while reflecting their spatial variation. This compromise leads to high accuracy in AD classification. We utilized MR volumes provided by Alzheimer's Disease Neuroimaging Initiative (ADNI) to validate the performance of the method. Our method discriminated AD patients from Healthy Control (HC) subjects with 82% sensitivity and 93% specificity. It also discriminated Mild Cognitive Impairment (MCI) patients, who converted to AD within 18 months, from non-converted MCI subjects with 63% sensitivity and 76% specificity. Moreover, it showed that the entorhinal cortex was the most discriminative region for classification, which is consistent with previous pathological findings. In comparison with other classification methods, our method demonstrated high classification performance in both categories, which supports the discriminative power of our method in both AD diagnosis and AD prediction.

Harmonization of Multicenter Cortical Thickness Data by Linear Mixed Effect Model.

Objective: Analyzing neuroimages being useful method in the field of neuroscience and neurology and solving the incompatibilities across protocols and vendors have become a major problem. We referred to this incompatibility as "center effects," and in this study, we attempted to correct such center effects of cortical feature obtained from multicenter magnetic resonance images (MRIs). Methods: For MRI of a total of 4,321 multicenter subjects, the harmonized w-score was calculated by correcting biological covariates such as age, sex, years of education, and intercranial volume (ICV) as fixed effects and center information as a random effect. Afterward, we performed classification tasks using principal component analysis (PCA) and linear discriminant analysis (LDA) to check whether the center effect was successfully corrected from the harmonized w-score. Results: First, an experiment was conducted to predict the dataset origin of a random subject sampled from two different datasets, and it was confirmed that the prediction accuracy of linear mixed effect (LME) model-based w-score was significantly closer to the baseline than that of raw cortical thickness. As a second experiment, we classified the data of the normal and patient groups of each dataset, and LME model-based w-score, which is biological-feature-corrected values, showed higher classification accuracy than the raw cortical thickness data. Afterward, to verify the compatibility of the dataset used for LME model training and the dataset that is not, intraobject comparison and w-score RMSE calculation process were performed. Conclusion: Through comparison between the LME model-based w-score and existing methods and several classification tasks, we showed that the LME model-based w-score sufficiently corrects the center effects while preserving the disease effects from the dataset. We also showed that the preserved disease effects have a match with well-known disease atrophy patterns such as Alzheimer's disease or Parkinson's disease. Finally, through intrasubject comparison, we found that the difference between centers decreases in the LME model-based w-score compared with the raw cortical thickness and thus showed that our model well-harmonizes the data that are not used for the model training.