Auto-Classification of Parkinson's Disease with Different Motor Subtypes Using Arterial Spin Labelling MRI Based on Machine Learning

MRI-based differentiation of Parkinson's disease by cerebellar gray matter volume

BACKGROUND

The underlying mechanism of Parkinson's disease (PD) is associated with the neurodegeneration of the dopaminergic neurons, and the cerebellum plays a significant role together in non-motor and motor functions in PD progression. Morphological changes in the cerebellum can greatly impact patients' clinical symptoms, especially motor control symptoms, and may also help distinguish patients from healthy subjects. This study aimed to explore the potential of cerebellar gray matter volume, related to motor control function, as a neuroimaging biomarker to classify patients with PD and healthy controls (HC) by using voxel-based morphometric (VBM) measurements and support vector machine (SVM) methods based on independent component analysis (ICA).

METHODS

Cerebellar gray matter volume was measured using VBM in patients with PD (n = 27) and HC (n = 16) from the Neurocon dataset. ICA analysis was performed on the gray matter volume of all subregions, resulting in 7 independent components. These independent components were then utilized for correlation analysis with clinical scales and trained as input features for the SVM model. PD patients (n = 20) and HC (n = 20) from the TaoWu dataset were used as test data to validate our SVM model.

RESULTS

Among patients with PD, 3 out of the 7 independent components showed a significant correlation with clinical scales. The SVM model achieved an accuracy of 86 % in classifying PD patients and HC, with a sensitivity of 72.2 %, specificity of 88 %, and F1 Score of 76.5 %. The accuracy of the SVM model verification analysis using the TaoWu dataset was 70 %, with a sensitivity of 62.5 %, a specificity of 100 %, and the F1 Score was 76.9 %.

CONCLUSIONS

The results suggest that abnormal cerebellar gray matter volume, which is highly correlated with motor control function in Parkinson's patients, may serve as a valuable neuroimaging biomarker capable of distinguishing Parkinson's patients from healthy individuals. We observed that the combination of the ICA method and the SVM method produced an improved classification model. This model may function as an early warning tool that enables clinicians to conduct preliminary identification and intervention for patients with PD.

Resting-State Functional MRI Regional Homogeneity Correlates With Motor Scores in Parkinson's Disease

BACKGROUND AND PURPOSE

This study investigated the neural mechanisms underlying Parkinson's disease (PD) subtypes-tremor dominant (TD) and postural instability gait difficulty (PIGD)-by analyzing regional homogeneity (ReHo) values from resting-state functional MRI.

METHODS

Fifty-nine PD patients (29 TD patients, 30 PIGD patients) and 30 healthy controls (HCs) were enrolled. ReHo values were analyzed via analysis of variance and a two-sample t-test, with age and sex as covariates. Correlations between ReHo values and clinical motor symptoms were also examined.

RESULTS

Distinct ReHo patterns were observed in patients with the PD subtypes and HCs. TD patients presented decreased ReHo in the cerebellar-thalamic-cortical circuit, whereas PIGD patients presented lower ReHo in the striatum and supplementary motor area (SMA). TD patients had higher ReHo in the bilateral dorsolateral superior frontal gyrus and SMA but lower ReHo in the bilateral medial orbital part of the superior frontal gyrus and other regions on the left than PIGD patients. Specific brain area ReHo values were correlated with tremor scores, PIGD scores, and rigidity scores.

CONCLUSION

Different motor subtypes of PD patients and HCs showed distinct ReHo patterns. ReHo correlation with clinical traits suggests its value as a biomarker for subtype-specific diagnostic strategies.

Causal associations between Helicobacter Pylori infection and the risk and symptoms of Parkinson's disease: a Mendelian randomization study

Background

Increasing evidence suggests an association between Helicobacter pylori (HP) infection and Parkinson's disease (PD) and its clinical manifestations, but the causal relationship remain largely unknown.

Objective

To investigate the causal relationship between HP infection and PD risk, PD symptoms, and secondary parkinsonism, we conducted two-sample Mendelian randomization (MR).

Methods

We obtained summary data from genome-wide association studies for seven different antibodies specific to HP proteins and five PD-related phenotypes. The inverse-variance weighted (IVW), weighted median, weighted mode, and MR-Egger methods were used to assess the causal relationships. Sensitivity analyses were performed to examine the stability of the MR results and reverse MR analysis was conducted to evaluate the presence of reverse causality.

Results

Genetically predicted HP antibodies were not causally associated with an increased risk of PD. However, HP cytotoxin-associated gene-A (CagA) and outer membrane protein (OMP) antibody level were causally associated with PD motor subtype (tremor to postural instability/gait difficulty score ratio; β = -0.16 and 0.46, P = 0.002 and 0.048, respectively). HP vacuolating cytotoxin-A (VacA) antibody level was causally associated with an increased risk of PD dementia [odds ratio (OR) = 1.93, P = 0.040]. Additionally, HP OMP antibody level was identified as a risk factor for drug-induced secondary parkinsonism (OR = 2.08, P = 0.033). These results were stable, showed no evidence of heterogeneity or directional pleiotropy, and no evidence of a reverse causal relationship.

Conclusions

Our findings indicate that HP infection does not increase the risk of PD, but contributes to PD motor and cognitive symptoms. Different types of HP antibodies affect different symptoms of PD. Eradication of HP infection may help modulate and improve symptoms in PD patients.

Multidimensional digital biomarker phenotypes for mild cognitive impairment: considerations for early identification, diagnosis and monitoring

Mild Cognitive Impairment (MCI) poses a challenge for a growing population worldwide. Early identification of risk for and diagnosis of MCI is critical to providing the right interventions at the right time. The paucity of reliable, valid, and scalable methods for predicting, diagnosing, and monitoring MCI with traditional biomarkers is noteworthy. Digital biomarkers hold new promise in understanding MCI. Identifying digital biomarkers specifically for MCI, however, is complex. The biomarker profile for MCI is expected to be multidimensional with multiple phenotypes based on different etiologies. Advanced methodological approaches, such as high-dimensional statistics and deep machine learning, will be needed to build these multidimensional digital biomarker profiles for MCI. Comparing patients to these MCI phenotypes in clinical practice can assist clinicians in better determining etiologies, some of which may be reversible, and developing more precise care plans. Key considerations in developing reliable multidimensional digital biomarker profiles specific to an MCI population are also explored.