Cognitive Decline in Parkinson's Disease: A Subgroup of Extreme Decliners Revealed by a Data-Driven Analysis of Longitudinal Progression

Machine learning for predicting cognitive decline within five years in Parkinson's disease: Comparing cognitive assessment scales with DAT SPECT and clinical biomarkers

OBJECTIVE

Parkinson's disease (PD) is an age-related neurodegenerative condition characterized mostly by motor symptoms. Although a wide range of non-motor symptoms (NMS) are frequently experienced by PD patients. One of the important and common NMS is cognitive impairment, which is measured using different cognitive scales. Monitoring cognitive impairment and its decline in PD is essential for patient care and management. In this study, our goal is to identify the most effective cognitive scale in predicting cognitive decline over a 5-year timeframe initializing clinical biomarkers and DAT SPECT.

METHODS

Machine Learning has previously shown superior performance in image and clinical data classification and detection. In this study, we propose to use machine learning with different types of data, such as DAT SPECT and clinical biomarkers, to predict PD-CD based on various cognitive scales. We collected 330 DAT SPECT images and their clinical data in baseline, years 2,3,4, and 5 from Parkinson's Progression Markers Initiative (PPMI). We then designed a 3D Autoencoder to extract deep radiomic features (DF) from DAT SPECT images, and we then concatenated it with 17 clinical features (CF) to predict cognitive decline based on Montreal Cognitive Assessment (MoCA) and The Movement Disorder Society-Unified Parkinson's Disease Rating Scale (MDS-UPDRS-I).

RESULTS

The utilization of MoCA as a cognitive decline scale yielded better performance in various years compared to MDS-UPDRS-I. In year 4, the application of the deep radiomic feature resulted in the highest achievement, with a cross-validation AUC of 89.28, utilizing the gradient boosting classifier. For the MDS-UPDRS-I scale, the highest achievement was obtained by utilizing the deep radiomic feature, resulting in a cross-validation AUC of 81.34 with the random forest classifier.

CONCLUSIONS

The study findings indicate that the MoCA scale may be a more effective predictor of cognitive decline within 5 years compared to MDS-UPDRS-I. Furthermore, deep radiomic features had better performance compared to sole clinical biomarkers or clinical and deep radiomic combined. These results suggest that using the MoCA score and deep radiomic features extracted from DAT SPECT could be a promising approach for identifying individuals at risk for cognitive decline in four years. Future research is needed to validate these findings and explore their utility in clinical practice.

Profiling people with Parkinson's disease at risk of cognitive decline: Insights from PPMI and ICICLE-PD data

Introduction

A subset of people with Parkinson's disease (PD) develop dementia faster than others. We aimed to profile PD cognitive subtypes at risk of dementia based on their rate of cognitive decline.

Method

Latent class mixed models stratified subtypes in Parkinson's Progression Markers Initiative (PPMI) (N = 770) and ICICLE-PD (N = 212) datasets based on their decline in the Montreal Cognitive Assessment over at least 4 years. Baseline demographic and cognitive data at diagnosis were compared between subtypes to determine their clinical profile.

Results

Four subtypes were identified: two with stable cognition, one with steady decline, and one with rapid decline. Performance on Judgement of Line Orientation, but not category fluency, was associated with a steady decline in the PPMI dataset, and deficits in category fluency, but not visuospatial function, were associated with a steady decline in the ICICLE-PD dataset.

Discussion

People with PD susceptible to cognitive decline demonstrate unique clinical profiles at diagnosis, although this differed between cohorts.

Highlights

Four cognitive subtypes were revealed in two Parkinson's disease samples.Unique profiles of cognitive impairment were related to cognitive decline.Judgement of Line Orientation/category fluency predictive of steady decline.Global deficits related to rapid cognitive decline and increased dementia risk.

Prediction of Cognitive Decline in Parkinson's Disease Using Clinical and DAT SPECT Imaging Features, and Hybrid Machine Learning Systems

BACKGROUND

We aimed to predict Montreal Cognitive Assessment (MoCA) scores in Parkinson's disease patients at year 4 using handcrafted radiomics (RF), deep (DF), and clinical (CF) features at year 0 (baseline) applied to hybrid machine learning systems (HMLSs).

METHODS

297 patients were selected from the Parkinson's Progressive Marker Initiative (PPMI) database. The standardized SERA radiomics software and a 3D encoder were employed to extract RFs and DFs from single-photon emission computed tomography (DAT-SPECT) images, respectively. The patients with MoCA scores over 26 were indicated as normal; otherwise, scores under 26 were indicated as abnormal. Moreover, we applied different combinations of feature sets to HMLSs, including the Analysis of Variance (ANOVA) feature selection, which was linked with eight classifiers, including Multi-Layer Perceptron (MLP), K-Neighbors Classifier (KNN), Extra Trees Classifier (ETC), and others. We employed 80% of the patients to select the best model in a 5-fold cross-validation process, and the remaining 20% were employed for hold-out testing.

RESULTS

For the sole usage of RFs and DFs, ANOVA and MLP resulted in averaged accuracies of 59 ± 3% and 65 ± 4% for 5-fold cross-validation, respectively, with hold-out testing accuracies of 59 ± 1% and 56 ± 2%, respectively. For sole CFs, a higher performance of 77 ± 8% for 5-fold cross-validation and a hold-out testing performance of 82 + 2% were obtained from ANOVA and ETC. RF+DF obtained a performance of 64 ± 7%, with a hold-out testing performance of 59 ± 2% through ANOVA and XGBC. Usage of CF+RF, CF+DF, and RF+DF+CF enabled the highest averaged accuracies of 78 ± 7%, 78 ± 9%, and 76 ± 8% for 5-fold cross-validation, and hold-out testing accuracies of 81 ± 2%, 82 ± 2%, and 83 ± 4%, respectively.

CONCLUSIONS

We demonstrated that CFs vitally contribute to predictive performance, and combining them with appropriate imaging features and HMLSs can result in the best prediction performance.

Systematic review of data-driven cognitive subtypes in Parkinson disease

BACKGROUND AND PURPOSE

Recent application of the mild cognitive impairment concept to Parkinson disease (PD) has proven valuable in identifying patients at risk of dementia. However, it has sparked controversy regarding the existence of cognitive subtypes. The present review evaluates the current literature pertaining to data-driven subtypes of cognition in PD.

METHODS

Following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, systematic literature searches for peer-reviewed articles on the topic of cognitive subtyping in PD were performed.

RESULTS

Twenty-two relevant articles were identified in the systematic search. Subtype structures showed either a spectrum of severity or specific domains of impairment. Domain-specific subtypes included amnestic/nonamnestic, memory/executive, and frontal/posterior dichotomies, as well as more complex structures with less definitive groupings. Preliminary longitudinal evidence showed some differences in cognitive progression among subtypes. Neuroimaging evidence provided insight into distinct patterns of brain alterations among subtypes.

CONCLUSIONS

Recurring phenotypes in the literature suggest strong clinical relevance of certain cognitive subtypes in PD. Although the current literature is limited, it raises critical questions about the utility of data-driven methods in cognitive research. The results encourage further integration of neuroimaging research to define the latent neural mechanisms behind divergent subtypes. Although there is no consensus, there appears to be growing consistency and inherent value in identifying cognitive subtypes in PD.

Multi-trajectory analysis uncovers latent associations between psychological and physiological acute stress response patterns

Encounter with an acute stressor elicits multiple physiological and psychological response trajectories that spread at different times-scales and directions. Associating a single physiological response trajectory with a specific psychological response has remained a challenge, due to putative interactions between the different stress response pathways. Hence, multidimensional analysis of stress response trajectories may be better suited to account for response variability. To test this, 96 healthy female participants underwent a robust acute laboratory stress induction procedure while their psychological [positive and negative affect (PANAS)] and physiological [heart rate (HR), heart rate variability (HRV), saliva cortisol (CORT)] responses were recorded before, during and after stress. Combining these data using unsupervised group-based multi-trajectory modelling uncovered three latent classes that best accounted for variability across psychological and physiological stress response trajectories. These classes were labelled based on their psychological response patterns as: A prototypical response group that depict a moderate increase in negative and decrease in positive affect during stress, with both patterns recovering after stress offset (n = 55); A heightened response group that depict excessive affective responses during stress that recover after stress offset (n = 24); and a lack of recovery group that depict a moderate increase in negative and decrease in positive affect during stress, with both patterns not recovering after stress offset (n = 17). With respect to physiological acute stress trajectories, all three groups exhibited comparable increases in HR and CORT during stress that recovered after stress offset, yet only the prototypical group expressed the expected stress-induced reduction in HRV, while the other two groups exhibited blunted HRV response. Critically, focusing on a single physiological stress response trajectory, including HRV, did not account for psychological response variability and vice versa. Taken together, a multi-trajectory approach may better account for the multidimensionality of acute stress response and uncover latent associations between psychological and physiological response patterns. Compared to the other two groups, the prototypical group also exhibited significantly lower overall stress scores based on the DASS-21 scale. This, alongside the uncovered response patterns, suggest that latent psycho-physiological associations may shed light on stress response adaptivity or lack thereof.

Morphological analysis of the brain subcortical gray structures in restless legs syndrome

BACKGROUND

Although several studies have shown the involvement of specific structures of the central nervous system, the dopaminergic system, and iron metabolism in restless legs syndrome (RLS), the exact location and extent of its anatomical substrate is not yet known. The scope of this new study was to investigate the brain subcortical gray structures, by means of structural magnetic resonance imaging (MRI) studies, in RLS patients in order to assess the presence of any volume or shape abnormalities involving these structures.

METHODS

Thirty-three normal controls (24 females and nine males) and 45 RLS patients (34 females and 11 males) were retrospectively recruited and underwent a 1.5 Tesla MRI study with two-dimensional T1 sequences in the sagittal plane. Post-processing was performed by means of the Functional Magnetic Resonance Imaging of the Brain Analysis Group Integrated Registration and Segmentation Tool (FIRST) software, and both volumetric and morphological analyses of the thalamus, caudate, putamen, globus pallidus, brainstem, hippocampus, and amygdala, bilaterally, were carried out.

RESULTS

A statistically significant volumetric reduction in the left amygdala and left globus pallidus was found in subjects with RLS, as well as large surface morphological alterations affecting the amygdala bilaterally and other less widespread surface changes in both hippocampi, the right caudate, the left globus pallidus, and the left putamen.

CONCLUSIONS

These findings seem to indicate that the basic mechanisms of RLS might include a pathway involving not only the hypothalamus-spinal dopaminergic circuit (nucleus A11), but also pathways including the basal ganglia and structures that are part of the limbic system; moreover, structural alterations in RLS seem to concern the morphology as well as the volume of the above structures. The role of basal ganglia in the complex neurophysiological and neurochemical mechanism of RLS needs to carefully reconsidered.