Abstract Background Patients with mild cognitive impairment(MCI) are a clinically important group because of the increased risk for the progression to Alzheimer’s dementia(AD). Amyloid positron emission tomography(PET) was useful as a predictor for identifying AD pathology in the pre‐dementia stage. We determined the rates of conversion to AD in MCI patients based on amyloid positivity by visual and quantitative analysis over 2 years. We also investigated the clinical and neuroimaging predictors for conversion to AD. Method Individuals aged 50 and above with MCI according to Petersen’s criteria were eligible. Subjects underwent laboratory tests including APOE genotyping, neuropsychological tests, brain magnetic resonance imaging(MRI), and amyloid PET assessed by 18 F‐FC119S. Amyloid positivity was determined in each lobe of the brain and quantitatively analyzed. The follow‐up period was at least 2 years. Cox regression analysis was performed to identify prognostic factors which are independently related to time to AD conversion. Result We recruited 50 patients and 39 subjects were enrolled. 5 patients dropped out at 2 years, so 34 subjects were finally included. 13 of 34(38.2%) by visual analysis and 12 of 34(35.3%) by quantitative analysis of 18 F‐FC119S PET showed amyloid positive. The basic demographics were not significantly different between the groups except for Korea‐Instrumental Activities of Daily Living(K‐IADL) at baseline(p = 0.004) and 1 year(p = 0.001), Mini‐mental State Examination(MMSE) at 2 years(p = 0.047) and Global Deterioration Scale(GDS) at 2 years(p = 0.029). 4 of 13(30.8%) by visual analysis, 4 of 12(66.7%) by quantitative analysis amyloid‐positive MCI patients converted to AD. Amyloid positive in posterior cingulate was marginally significant(p = 0.066). AD conversion group had a higher Clinical Dementia Rating‐Sum of box(CDR‐SOB) at baseline(p = 0.002) compared with remained MCI group. In multivariate cox regression, CDR‐SOB at baseline, MMSE at 1 year and K‐IADL at 2 years were included. Conclusion Amyloid‐positive MCI group was more likely to progress to AD. But long‐term follow‐up was required. By quantitative analysis of 18 F‐FC119S PET, amyloid deposition in posterior cingulate was highly associated with the AD conversion group. CDR‐SOB at baseline, MMSE at 1 year and K‐IADL at 2 years were considered as clinically valuable prognostic predictors of disease progression. Especially, CDR‐SOB is important as a predictive biomarker in the pre‐dementia stage.

Abstract Background Early screening of dementia at its pre‐clinical stage is crucial, seeing as how deteriorated cognitive functions can be recovered through appropriate medicinal practices. The pre‐clinical stage, which we refer to as mild cognitive impairment (MCI) is known to affect the alpha rhythm of the brain activity. Hence, our study aims to establish quantitative EEG (qEEG)‐based alpha asymmetry biomarker that aids the diagnoses of MCI. Method The qEEG dataset were composed of age‐ and sex‐ matched subjects (N = 634, 317 healthy control (HC), 317 MCI), acquired in accordance with the 10‐20 system and under resting state with their eyes being closed. Subjects were recruited by Chung‐Ang University (153 HC, 111 MCI), iMediSync(164 HC, 76 MCI), Dong‐a Medical Center(84 MCI) and Pusan Medical Center(46 MCI). The features were classified into four types: the absolute and relative region alpha power (AAP/RAP) of six regions (Prefrontal, Frontal, Temporal, Central, Parietal, Occipital); alpha power asymmetry (APA) between regions that align in anterior‐posterior and lateral planes; occipital alpha peak frequency (OPF) that analyzes peak frequencies. In order to highlight the alpha variability, EEG signals were bandpass filtered (6.5‐12Hz). Filtered signal at each channel was categorized into slow (6.5‐8Hz), middle (8‐10Hz) and fast alpha (10‐12Hz). The mean and standard deviation of their FFT were computed through moving time window. Henceforth, the feature set were used to train machine learning algorithms. Data were split into 8(training N = 507) to 2(test N = 127) ratio. 101 subjects were randomly selected and used as validation set. The algorithm that exhibited the best classification performance were determined through 5‐fold cross validation (CV) results. Result Support vector machine (SVM) showed the best 5‐fold CV results – 84.61% average accuracy, 74.02% sensitivity, 95.25% specificity, 94.15% precision and F1‐Score of 0.8273. The RAP decreased for all regions and OPF were slower in MCI group when compared to HC. APA patterns of the right intra‐hemisphere exhibited most significant differences. Conclusion This study illustrates the potential of APA as a biomarker for MCI. Current applications of APA are confined to emotional‐related studies, such as depressive disorder. The reported cases of MCI‐associated depression also uphold the validity of APA as a biomarker for MCI.

Abstract Background Early phase amyloid positron emission tomography (PET) imaging is known to correlated with glucose metabolism. However, studies on the optimal time frame and SUVR cut off for early‐phase amyloid PET have not been sufficiently progressed. In this study, the optimal time of early‐phase 18F‐FBB was determined and SUVR cut off was obtained. Method To determine the optimal time frame, the correlation between FBB and FDG SUVR of 67 subjects(34, Alzheimer’s disease(AD), 4 mild cognitive impairment(MCI), and 29 health control(HC)) was compared. In this method, 83 brain regions were treated as gray matter masks using an integrated brain map of PMOD 3.6. After obtaining the optimal time of FBB, which has a high correlation compared to FDG, a total of 225 subjects(188 AD, 37 HC) were targeted. SUVR cut off of dual‐phase 18F‐FBB PET of 11 brain regions was obtained. Calculate the composite mean SUVR and regional SUVR using MedCalc. Result The highest correlation between early‐phase 18F‐FBB and 18F‐FDG peaked at ‐.8747 at 270‐330s. Considering the minimum time and correlation between image noise and visual analysis, 90 to 330 seconds were set as the optimal time frame(r‐ratio ≥0.875). In the early‐phase FBB, cut off values in the order of HC>AD were valid. The classification accuracy of the composite mean SUVR was 72%(AD 69.15%, HC 86.49%), and the regional area accuracy was 85.78%(AD 93.62%, HC 45.95%). In the delay‐phase FBB, cut off vlaues in the order of AD>HC were valid. The classification accuracy of the composite mean SUVR was 77,33%(AD 72,87%, HC 100%), and the regional area accuracy was 84.89%(AD 86.17%, HC 78.38%). In the early and delay‐phase FBB, the classification accuracy of the composite mean SUVR was 90.67%(AD 92.02%, HC 86.49%), and the regional area accuracy was 88%(AD 98.4%, HC 37.84%). Conclusion In this study, it is possible to know the optimal time frame of the early‐phase 18F‐FBB showing a high correlation with 18F‐FDG. Composite mean SUVR and regional SUVR cut off of the dual‐phase 18F‐FBB are helpful for visually ambiguous analysis.

Alzheimer’s disease (AD) and Parkinson’s disease (PD) are prevalent neurodegenerative disorders among the elderly, leading to cognitive decline and motor impairments. As the population ages, the prevalence of these neurodegenerative disorders is increasing, providing motivation for active research in this area. However, most studies are conducted using brain imaging, with relatively few studies utilizing voice data. Using voice data offers advantages in accessibility compared to brain imaging analysis. This study introduces a novel ensemble-based classification model that utilizes Mel spectrograms and Convolutional Neural Networks (CNNs) to distinguish between healthy individuals (NM), AD, and PD patients. A total of 700 voice samples were collected under standardized conditions, ensuring data reliability and diversity. The proposed ternary classification algorithm integrates the predictions of binary CNN classifiers through a majority voting ensemble strategy. ResNet, DenseNet, and EfficientNet architectures were employed for model development. The experimental results show that the ensemble model based on ResNet achieves a weighted F1 score of 91.31%, demonstrating superior performance compared to existing approaches. To the best of our knowledge, this is the first large-scale study to perform three-class classification of neurodegenerative diseases using voice data.

In amyloid brain PET, after parcellation using the finite element method (FEM)-based algorithm FreeSurfer and voxel-based algorithm PMOD, SUVr examples can be extracted and compared. This study presents the classification SUVr threshold in PET images of F-18 florbetaben (FBB), F-18 flutemetamol (FMM), and F-18 florapronol (FPN) and compares and analyzes the classification performance according to computational algorithm in each brain region. PET images were co-registered after the generated MRI was registered with standard template information. Using MATLAB script, SUVr was calculated using the built-in parcellation number labeled in the brain region. PMOD and FreeSurfer with different algorithms were used to load the PET image, and after registration in MRI, it was normalized to the MRI template. The volume and SUVr of the individual gray matter space region were calculated using an automated anatomical labeling atlas. The SUVr values of eight regions of the frontal cortex (FC), lateral temporal cortex (LTC), mesial temporal cortex (MTC), parietal cortex (PC), occipital cortex (OC), anterior and posterior cingulate cortex (GCA, GCP), and composite were calculated. After calculating the correlation of SUVr using the FreeSurfer and PMOD algorithms and calculating the AUC for amyloid-positive/negative subjects, the classification ability was calculated, and the SVUr threshold was calculated using the Youden index. The correlation coefficients of FreeSurfer and PMOD SUVr calculations of the eight regions of the brain cortex were FBB (0.95), FMM (0.94), and FPN (0.91). The SUVr threshold was SUVr(LTC,min) = 1.264 and SUVr(THA,max) = 1.725 when calculated using FPN-FreeSurfer, and SUVr(MTC,min) = 1.093 and SUVr(MCT,max) = 1.564 when calculated using FPN-PMOD. The AUC comparison showed that there was no statistically significant difference (p > 0.05) in the SUVr classification results using the three radiopharmaceuticals, specifically for the LTC and OC regions in the PMOD analysis, and the LTC and PC regions in the FreeSurfer analysis. The SUVr calculation using PMOD (voxel-based algorithm) has a strong correlation with the calculation using FreeSurfer (FEM-based algorithm); therefore, they complement each other. Quantitative classification analysis with high accuracy is possible using the suggested SUVr threshold. The SUVr classification performance was good in the order of FMM, FBB, and FPN, and showed a good classification performance in the LTC region regardless of the type of radiotracer and analysis algorithm.

In amyloid brain PET, after parcellation using the finite element method (FEM)-based algorithm FreeSurfer and voxel-based algorithm PMOD, SUVr examples can be extracted and compared. This study presents the classification SUVr threshold in PET images of F-18 florbetaben (FBB), F-18 flutemetamol (FMM), and F-18 florapronol (FPN) and compares and analyzes the classifica-tion performance according to computational algorithm in each brain region. PET images were co-registered after the generated MRI was registered with standard template information. Using MATLAB script, SUVr was calculated using the built-in parcellation number labeled in the brain region. PMOD and FreeSurfer with different algorithms were used to load the PET image, and after registration in MRI, it was normalized to the MRI template. The volume and SUVr of the individual gray matter space region were calculated using an automated anatomical labeling at-las. The SUVr values of eight regions of the frontal cortex (FC), lateral temporal cortex (LTC), mesial temporal cortex (MTC), parietal cortex (PC), occipital cortex (OC), anterior and posterior cingulate cortex (GCA, GCP), and composite were calculated. After calculating the correlation of SUVr using the FreeSurfer and PMOD algorithms and calculating the AUC, the classification ability was calculated, and the SVUr threshold was calculated using the Youden index. The cor-relation coefficients of FreeSurfer and PMOD SUVr calculations of the eight regions of the brain cortex were FBB (0.95), FMM (0.94), and FPN (0.91). The SUVr threshold was SUVr(LTC,min) = 1.264 and SUVr(THA,max) = 1.725 when calculated using FPN-FreeSurfer, and SUVr(MTC,min) = 1.093 and SUVr(MCT,max) = 1.564 when calculated using FPN-PMOD. The AUC comparison showed that there was no statistically significant difference (p>0.05) in SUVr classification results using the three radiopharmaceuticals, specifically for LTC and OC regions in PMOD analysis, and LTC and PC regions in FreeSurfer analysis. The SUVr calculation using PMOD (voxel-based algorithm) has a strong correlation with the calculation using FreeSurfer (FEM-based algorithm); therefore, they are complement with each other. Quantitative classification analysis with high accuracy is possible using the suggested SUVr threshold. The SUVr classification performance was good in the order of FMM, FBB, and FPN and showed good classification performance in the LTC region, regardless of the type of radiotracer and analysis algorithm.

Alzheimer's disease (AD), the most prevalent degenerative brain disease associated with dementia, requires early diagnosis to alleviate worsening of symptoms through appropriate management and treatment. Recent studies on AD stage classification are increasingly using multimodal data. However, few studies have applied graph neural networks to multimodal data comprising F-18 florbetaben (FBB) amyloid brain positron emission tomography (PET) images and clinical indicators. The objective of this study was to demonstrate the effectiveness of graph convolutional network (GCN) for AD stage classification using multimodal data, specifically FBB PET images and clinical indicators, collected from Dong-A University Hospital (DAUH) and Alzheimer's Disease Neuroimaging Initiative (ADNI). The effectiveness of GCN was demonstrated through comparisons with the support vector machine, random forest, and multilayer perceptron across four classification tasks (normal control (NC) vs. AD, NC vs. mild cognitive impairment (MCI), MCI vs. AD, and NC vs. MCI vs. AD). As input, all models received the same combined feature vectors, created by concatenating the PET imaging feature vectors extracted by the 3D dense convolutional network and non-imaging feature vectors consisting of clinical indicators using multimodal feature fusion method. An adjacency matrix for the population graph was constructed using cosine similarity or the Euclidean distance between subjects' PET imaging feature vectors and/or non-imaging feature vectors. The usage ratio of these different modal data and edge assignment threshold were tuned by setting them as hyperparameters. In this study, GCN-CS-com and GCN-ED-com were the GCN models that received the adjacency matrix constructed using cosine similarity (CS) and the Euclidean distance (ED) between the subjects' PET imaging feature vectors and non-imaging feature vectors, respectively. In modified nested cross validation, GCN-CS-com and GCN-ED-com respectively achieved average test accuracies of 98.40%, 94.58%, 94.01%, 82.63% and 99.68%, 93.82%, 93.88%, 90.43% for the four aforementioned classification tasks using DAUH dataset, outperforming the other models. Furthermore, GCN-CS-com and GCN-ED-com respectively achieved average test accuracies of 76.16% and 90.11% for NC vs. MCI vs. AD classification using ADNI dataset, outperforming the other models. These results demonstrate that GCN could be an effective model for AD stage classification using multimodal data.

Patients with mild cognitive impairment (MCI) have a relatively high risk of developing Alzheimer's dementia (AD), so early identification of the risk for AD conversion can lessen the socioeconomic burden. In this study, ¹⁸F-Florapronol, newly developed in Korea, was used for qualitative and quantitative analyses to assess amyloid positivity. We also investigated the clinical predictors of the progression from MCI to dementia over 2 years. From December 2019 to December 2022, 50 patients with MCI were recruited at a single center, and 34 patients were included finally. Based on visual analysis, 13 (38.2%) of 34 participants were amyloid-positive, and 12 (35.3%) were positive by quantitative analysis. Moreover, 6 of 34 participants (17.6%) converted to dementia after a 2-year follow-up (<i>p</i> = 0.173). Among the 15 participants who were positive for amyloid in the posterior cingulate region, 5 (33.3%) patients developed dementia (<i>p</i> = 0.066). The Clinical Dementia Rating-Sum of Boxes (CDR-SOB) at baseline was significantly associated with AD conversion in multivariate Cox regression analyses (<i>p</i> = 0.043). In conclusion, these results suggest that amyloid positivity in the posterior cingulate region and higher CDR-SOB scores at baseline can be useful predictors of AD conversion in patients with MCI.

Abstract Background Patients with mild cognitive impairment(MCI) are a clinically important group because of the increased risk for the progression to Alzheimer’s dementia(AD). Amyloid positron emission tomography(PET) was useful as a predictor for identifying AD pathology in the pre‐dementia stage. We determined the rates of conversion to AD in MCI patients based on amyloid positivity by visual and quantitative analysis over 2 years. We also investigated the clinical and neuroimaging predictors for conversion to AD. Method Individuals aged 50 and above with MCI according to Petersen’s criteria were eligible. Subjects underwent laboratory tests including APOE genotyping, neuropsychological tests, brain magnetic resonance imaging(MRI), and amyloid PET assessed by 18 F‐FC119S. Amyloid positivity was determined in each lobe of the brain and quantitatively analyzed. The follow‐up period was at least 2 years. Cox regression analysis was performed to identify prognostic factors which are independently related to time to AD conversion. Result We recruited 50 patients and 39 subjects were enrolled. 5 patients dropped out at 2 years, so 34 subjects were finally included. 13 of 34(38.2%) by visual analysis and 12 of 34(35.3%) by quantitative analysis of 18 F‐FC119S PET showed amyloid positive. The basic demographics were not significantly different between the groups except for Korea‐Instrumental Activities of Daily Living(K‐IADL) at baseline(p = 0.004) and 1 year(p = 0.001), Mini‐mental State Examination(MMSE) at 2 years(p = 0.047) and Global Deterioration Scale(GDS) at 2 years(p = 0.029). 4 of 13(30.8%) by visual analysis, 4 of 12(66.7%) by quantitative analysis amyloid‐positive MCI patients converted to AD. Amyloid positive in posterior cingulate was marginally significant(p = 0.066). AD conversion group had a higher Clinical Dementia Rating‐Sum of box(CDR‐SOB) at baseline(p = 0.002) compared with remained MCI group. In multivariate cox regression, CDR‐SOB at baseline, MMSE at 1 year and K‐IADL at 2 years were included. Conclusion Amyloid‐positive MCI group was more likely to progress to AD. But long‐term follow‐up was required. By quantitative analysis of 18 F‐FC119S PET, amyloid deposition in posterior cingulate was highly associated with the AD conversion group. CDR‐SOB at baseline, MMSE at 1 year and K‐IADL at 2 years were considered as clinically valuable prognostic predictors of disease progression. Especially, CDR‐SOB is important as a predictive biomarker in the pre‐dementia stage.

Abstract Background Early screening of dementia at its pre‐clinical stage is crucial, seeing as how deteriorated cognitive functions can be recovered through appropriate medicinal practices. The pre‐clinical stage, which we refer to as mild cognitive impairment (MCI) is known to affect the alpha rhythm of the brain activity. Hence, our study aims to establish quantitative EEG (qEEG)‐based alpha asymmetry biomarker that aids the diagnoses of MCI. Method The qEEG dataset were composed of age‐ and sex‐ matched subjects (N = 634, 317 healthy control (HC), 317 MCI), acquired in accordance with the 10‐20 system and under resting state with their eyes being closed. Subjects were recruited by Chung‐Ang University (153 HC, 111 MCI), iMediSync(164 HC, 76 MCI), Dong‐a Medical Center(84 MCI) and Pusan Medical Center(46 MCI). The features were classified into four types: the absolute and relative region alpha power (AAP/RAP) of six regions (Prefrontal, Frontal, Temporal, Central, Parietal, Occipital); alpha power asymmetry (APA) between regions that align in anterior‐posterior and lateral planes; occipital alpha peak frequency (OPF) that analyzes peak frequencies. In order to highlight the alpha variability, EEG signals were bandpass filtered (6.5‐12Hz). Filtered signal at each channel was categorized into slow (6.5‐8Hz), middle (8‐10Hz) and fast alpha (10‐12Hz). The mean and standard deviation of their FFT were computed through moving time window. Henceforth, the feature set were used to train machine learning algorithms. Data were split into 8(training N = 507) to 2(test N = 127) ratio. 101 subjects were randomly selected and used as validation set. The algorithm that exhibited the best classification performance were determined through 5‐fold cross validation (CV) results. Result Support vector machine (SVM) showed the best 5‐fold CV results – 84.61% average accuracy, 74.02% sensitivity, 95.25% specificity, 94.15% precision and F1‐Score of 0.8273. The RAP decreased for all regions and OPF were slower in MCI group when compared to HC. APA patterns of the right intra‐hemisphere exhibited most significant differences. Conclusion This study illustrates the potential of APA as a biomarker for MCI. Current applications of APA are confined to emotional‐related studies, such as depressive disorder. The reported cases of MCI‐associated depression also uphold the validity of APA as a biomarker for MCI.