Explore the vast range of titles available.

BackgroundSleep disorders can occur in early Parkinson’s disease (PD). However, the relationship between different sleep disturbances and their longitudinal evolution has not been fully explored.ObjectiveTo describe the frequency, coexistence, and longitudinal change in excessive daytime sleepiness (EDS), insomnia, and probable REM sleep behavior disorder (pRBD) in early PD.MethodsData were obtained from the Parkinson’s Progression Markers Initiative (PPMI). EDS, insomnia, and pRBD were defined using the Epworth Sleepiness Scale, MDS-UPDRS Part I sub-item 1.7, and RBD screening questionnaire.Results218 PD subjects and 102 controls completed 5 years of follow-up. At baseline, 69 (31.7%) PD subjects reported one type of sleep disturbance, 25 (11.5%) reported two types of sleep disturbances, and three (1.4%) reported all three types of sleep disturbances. At 5 years, the number of PD subjects reporting one, two, and three types of sleep disturbances was 85 (39.0%), 51 (23.4%), and 16 (7.3%), respectively. Only 41(18.8%) patients were taking sleep medications. The largest increase in frequency was seen in insomnia (44.5%), followed by EDS (32.1%) and pRBD (31.2%). Insomnia was the most common sleep problem at any time over the 5-year follow-up. The frequency of sleep disturbances in HCs remained stable.ConclusionsThere is a progressive increase in the frequency of sleep disturbances in PD, with the number of subjects reporting multiple sleep disturbances increasing over time. Relatively a few patients reported multiple sleep disturbances, suggesting that they can have different pathogenesis. A large number of patients were not treated for their sleep disturbances.

IntroductionDiscrete patterns of progression have been suggested for patients with Parkinson disease and presenting tremor dominant (TD) or postural instability gait disorders (PIGD). However, longitudinal prospective assessments need to take into consideration the variability in clinical manifestations and the evidence that only 40% of initially classified PIGD remain in this subtype at subsequent visits.MethodsWe analyzed clinical progression of PIGD compared to TD using longitudinal clinical data from the PPMI. Given the reported instability of such clinical classification, we only included patients who were reported as PIGD/TD at each visit during the 4-year observation. We used linear mixed-effects models to test differences in progression in these subgroups in 51 dependent variables.ResultsThere were 254 patients with yearly assessment. The number of PIGD was 36/254 vs 144/254 TD. PIGD had more severe motor disease at baseline but progressed faster than TD only in three non-motor items of the MDS-UPDRS: cognitive impairment, hallucinations, and psychosis plus features of DDS. Our analysis also showed in PIGD faster increase in the average time with dyskinesia.ConclusionsPIGD are characterized by more severe disease manifestations at diagnosis and greater cognitive progression, more frequent hallucinations, psychosis as well as features of DDS than TD patients. We interpret these findings as expression of greater cortical and subcortical involvement in PIGD already at onset. Since PIGD/TD classification is very unstable at onset, our analysis based on stricter definition criteria provides important insight for clinical trial stratification and definition of related outcome measures.

Background With the development of biotechnology and the accumulation of theories, many studies have found that microRNAs (miRNAs) play an important role in various diseases. Uncovering the potential associations between miRNAs and diseases is helpful to better understand the pathogenesis of complex diseases. However, traditional biological experiments are expensive and time-consuming. Therefore, it is necessary to develop more efficient computational methods for exploring underlying disease-related miRNAs. Results In this paper, we present a new computational method based on positive point-wise mutual information (PPMI) and attention network to predict miRNA-disease associations (MDAs), called PATMDA. Firstly, we construct the heterogeneous MDA network and multiple similarity networks of miRNAs and diseases. Secondly, we respectively perform random walk with restart and PPMI on different similarity network views to get multi-order proximity features and then obtain high-order proximity representations of miRNAs and diseases by applying the convolutional neural network to fuse the learned proximity features. Then, we design an attention network with neural aggregation to integrate the representations of a node and its heterogeneous neighbor nodes according to the MDA network. Finally, an inner product decoder is adopted to calculate the relationship scores between miRNAs and diseases. Conclusions PATMDA achieves superior performance over the six state-of-the-art methods with the area under the receiver operating characteristic curve of 0.933 and 0.946 on the HMDD v2.0 and HMDD v3.2 datasets, respectively. The case studies further demonstrate the validity of PATMDA for discovering novel disease-associated miRNAs.

PurposeThis work aimed to assess the potential of a set of features extracted from [123I]FP-CIT SPECT brain images to be used in the computer-aided “in vivo” confirmation of dopaminergic degeneration and therefore to assist clinical decision to diagnose Parkinson’s disease.MethodsSeven features were computed from each brain hemisphere: five standard features related to uptake ratios on the striatum and two features related to the estimated volume and length of the striatal region with normal uptake. The features were tested on a dataset of 652 [123I]FP-CIT SPECT brain images from the Parkinson’s Progression Markers Initiative. The discrimination capacities of each feature individually and groups of features were assessed using three different machine learning techniques: support vector machines (SVM), k-nearest neighbors and logistic regression.ResultsCross-validation results based on SVM have shown that, individually, the features that generated the highest accuracies were the length of the striatal region (96.5%), the putaminal binding potential (95.4%) and the striatal binding potential (93.9%) with no statistically significant differences among them. The highest classification accuracy was obtained using all features simultaneously (accuracy 97.9%, sensitivity 98% and specificity 97.6%). Generally, slightly better results were obtained using the SVM with no statistically significant difference to the other classifiers for most of the features.ConclusionsThe length of the striatal region uptake is clinically useful and highly valuable to confirm dopaminergic degeneration “in vivo” as an aid to the diagnosis of Parkinson’s disease. It compares fairly well to the standard uptake ratio-based features, reaching, at least, similar accuracies and is easier to obtain automatically. Thus, we propose its day to day clinical use, jointly with the uptake ratio-based features, in the computer-aided diagnosis of dopaminergic degeneration in Parkinson’s disease.

The accumulation of beta amyloid in the brain has a complex and poorly understood impact on the progression of Parkinson’s disease pathology and much controversy remains regarding its role, specifically in cognitive decline symptoms. Some studies have found increased beta amyloid burden is associated with worsening cognitive impairment in Parkinson’s disease, especially in cases where dementia occurs, while other studies failed to replicate this finding. To better understand this relationship, we examined a cohort of 25 idiopathic Parkinson’s disease patients and 30 healthy controls from the Parkinson’s Progression Marker Initiative database. These participants underwent [ 18 F]Florbetaben positron emission tomography scans to quantify beta amyloid deposition in 20 cortical regions. We then analyzed this beta amyloid data alongside the longitudinal Montreal Cognitive Assessment scores across 3 years to see how participant’s baseline beta amyloid levels affected their cognitive scores prospectively. The first analysis we performed with these data was a hierarchical cluster analysis to help identify brain regions that shared similarity. We found that beta amyloid clusters differently in Parkinson’s disease patients compared to healthy controls. In the Parkinson’s disease group, increased beta amyloid burden in cluster 2 was associated with worse cognitive ability, compared to deposition in clusters 1 or 3. We also performed a stepwise linear regression where we found an adjusted R 2 of 0.495 (49.5%) in a model explaining the Parkinson’s disease group’s Montreal Cognitive Assessment score 1-year post-scan, encompassing the left gyrus rectus, the left anterior cingulate cortex, and the right parietal cortex. Taken together, these results suggest regional beta amyloid deposition alone has a moderate effect on predicting future cognitive decline in Parkinson’s disease patients. The patchwork effect of beta amyloid deposition on cognitive ability may be part of what separates cognitive impairment from cognitive sparing in Parkinson’s disease. Thus, we suggest it would be more useful to measure beta amyloid burden in specific brain regions rather than using a whole-brain global beta amyloid composite score and use this information as a tool for determining which Parkinson’s disease patients are most at risk for future cognitive decline.

The correlation between serum sodium and sporadic Parkinson's disease remains unclear currently. This study aimed to assess the association between serum sodium and sporadic Parkinson's disease. The ultimate goal is to gain a deeper understanding of the implications of this relationship between serum sodium and sporadic Parkinson's disease. We conducted a retrospective cross-sectional study involving 1,189 participants in PPMI cohort. Age, sex, education years, race, body mass index, calcium, alanine aminotransferase, aspartate aminotransferase, white blood cell, lymphocytes, neutrophils, monocytes, red blood cell, hemoglobin, platelets, total protein, albumin, serum uric acid, serum sodium, serum potassium, urea nitrogen, creatinine, serum glucose were obtained from all participants. Logistic regression, and smooth curve fitting were utilized to substantiate the research objectives. The overall sporadic Parkinson's disease was 77.5% (921/1189); it was 71.9% (143/199), 75.4% (295/391), 76.7% (171/223), and 83% (312/376) for serum sodium quantile1 (Q1, 130-138.9 mmol/L), quantile 2 (Q2, 139-140.9 mmol/L), quantile 3 (Q3, 141-141.9 mmol/L), and quantile 4 (Q4, 142-155 mmol/L), respectively (  = 0.011). Multivariate odds ratio regression adjusted for risk factors demonstrates a 1-unit increment in the serum sodium raises the risk of sporadic Parkinson's disease by 1.11 times, respectively. Smooth splines analysis suggested a linear association between levels of serum sodium and risk of sporadic Parkinson's disease (P nonlinearity = 0.5). An interaction was observed between serum sodium and sex in their influence on sporadic Parkinson's disease (  < 0.05). Further exploratory subgroup analysis within the age and BMI groups showed that there were no significant interactions between the subgroups (all values for interaction were > 0.05). Additional sensitivity analyses supported the primary findings and indicated the conclusions are robust. This study highlights the influence of inappropriate serum sodium on the risk of incident sporadic Parkinson's disease, independent of confounders. The link between serum sodium and sporadic Parkinson's disease is linear.

Drug combination therapies are a promising strategy to overcome drug resistance and improve the efficacy of monotherapy in cancer, and it has been shown to lead to a decrease in dose-related toxicities. Except the synergistic reaction between drugs, some antagonistic drug–drug interactions (DDIs) exist, which is the main cause of adverse drug events. Precisely predicting the type of DDI is important for both drug development and more effective drug combination therapy applications. Recently, numerous text mining– and machine learning–based methods have been developed for predicting DDIs. All these methods implicitly utilize the feature of drugs from diverse drug-related properties. However, how to integrate these features more efficiently and improve the accuracy of classification is still a challenge. In this paper, we proposed a novel method (called NMDADNN) to predict the DDI types by integrating five drug-related heterogeneous information sources to extract the unified drug mapping features. NMDADNN first constructs the similarity networks by using the Jaccard coefficient and then implements random walk with restart algorithm and positive pointwise mutual information for extracting the topological similarities. After that, five network-based similarities are unified by using a multimodel deep autoencoder. Finally, NMDADNN implements the deep neural network (DNN) on the unified drug feature to infer the types of DDIs. In comparison with other recent state-of-the-art DNN-based methods, NMDADNN achieves the best results in terms of accuracy, area under the precision-recall curve, area under the ROC curve, F1 score, precision and recall. In addition, many of the promising types of drug–drug pairs predicted by NMDADNN are also confirmed by using the interactions checker tool. These results demonstrate the effectiveness of our NMDADNN method, indicating that NMDADNN has the great potential for predicting DDI types.

Still image-based human action recognition (HAR) is one of the most challenging research problems in the field of computer vision. Some of the significant reasons to support this claim are the availability of few datasets as well as fewer images per action class and the existence of many confusing classes in the datasets and comparing with video-based data. There is the unavailability of temporal information. In this work, we train some of the most reputed Convolutional Neural Network (CNN) based architectures using transfer learning after fine-tuned those suitably to develop a model for still image-based HAR. Since the number of images per action classes is found to be significantly less in number, we have also applied some well-known data augmentation techniques to increase the amount of data, which is always a need for deep learning-based models. Two benchmark datasets used for validating our model are Stanford 40 and PPMI, which are better known for their confusing action classes and the presence of occluded images and random poses of subjects. Results obtained by our model on these datasets outperform some of the benchmark results reported in the literature by a considerable margin. Class imbalance is deliberately introduced in the said datasets to better explore the robustness of the proposed model. The source code of the present work is available at: https://github.com/saikat021/Transfer-Learning-Based-HAR

IntroductionSeveral genetic variants are associated with an increased risk for developing Parkinson’s Disease (PD) and limited genotype/phenotype correlation. Specifically, mutations in GBA1, the gene coding for the lysosomal enzyme glucocerebrosidase, are associated with an earlier age of onset and faster disease progression. Given these phenotypic differences associated with GBA1 variants, we explored whether cortical thickness and other biomarkers of neurodegeneration differed in healthy controls and PD patients with and without GBA1 variants.MethodsTo understand how different GBA1 variants influence PD phenotype early in the disease, we retrieved neuroimaging and biospecimen data from the Parkinson’s Progression Markers Initiative database. Using FreeSurfer, we compared T1-weighted MRI images from healthy controls (N = 47) to PD patients with heterozygous N370S (N = 21), heterozygous E326K (N = 18) or heterozygous T369M (N = 8) variants, and GBA1 non-mutation carriers (N = 47).ResultsCortical thickness in PD patients differed from controls in the parietal cortex, with E365K, T369M variants, and GBA1 non-mutation carriers showing more cortical thinning than N370S variants. Patients with N370S variants had significantly higher serum neurofilament light levels among all groups.ConclusionOur results demonstrate significant cortical thinning in PD patients independent of genotype in superior parietal and postcentral regions when compared to the controls. They highlight the impact of GBA1 variants on cortical thickness in the parietal cortex. Finally, they suggest that recently diagnosed PD patients with N370S variants have a higher cortical thickness and increased active neurodegeneration when compared to PD patients without GBA1 mutations and PD patients with E326K or T369M variants.

Parkinson’s disease (PD) is a progressive neurodegenerative disorder shaped by genetic factors such as LRRK2 and GBA1 mutations, as well as dietary and metabolic influences. Food-derived plasma metabolites—including caffeine, paraxanthine, trigonelline, piperine, and sitosteryl hexoside—have emerged as promising, accessible biomarkers for early detection, progression monitoring, and therapeutic targeting, yet their longitudinal behavior and genetic interactions remain insufficiently characterized. Using the Parkinson’s Progression Markers Initiative (PPMI) cohort (n = 455; 303 PD patients, 152 controls), we quantified plasma levels of these metabolites by quantitative LC-MS/MS with batch correction, examining sporadic PD and genetically defined subgroups (LRRK2-PD [PDL], GBA1-PD [PDG], dual-mutation PD [PDGL], and prodromal equivalents). Baseline one-way ANOVA showed significantly lower caffeine and paraxanthine in PDL (p = 0.0467, p = 0.0178) and PDG (p = 0.0408), reduced piperine in PDL (p = 0.0009), PDG (p = 0.0257), and prodromal LRRK2 (p = 0.0168), and elevated sitosteryl hexoside in PDG (p = 0.0184). Longitudinal regression analyses revealed that in sporadic PD, caffeine negatively correlated with MDS-UPDRS parts I (β = −2, p = 0.0475) and III (β = −7.2, p = 0.007), trigonelline declined over time and was inversely associated with part III (β = −1.7, p = 0.0069), and sitosteryl hexoside negatively correlated with parts II (β = −68.3, p = 0.042) and III (β = −74.1, p = 0.0425). In PDL, sitosteryl hexoside inversely correlated with part I (β = −54.2, p = 0.0049), while in PDGL, paraxanthine showed negative associations with part II (β = −18.5, p = 0.00327). These findings demonstrate subgroup-specific alterations in food-derived metabolites and consistent inverse associations with PD severity, supporting their potential as non-invasive biomarkers, particularly in LRRK2/GBA1 mutation carriers, and highlighting the need for longitudinal validation and dietary intervention trials to advance personalized PD management.