Predicting rapid progression of Parkinson's Disease at baseline patients evaluation

UPSIT subitems may predict motor progression in Parkinson's disease

Background

The relationship between hyposmia and motor progression is controversial in Parkinson's disease (PD). The aim of this study was to investigate whether preserved identification of Chinese-validated University of Pennsylvania Smell Identification Test (UPSIT) odors could predict PD motor progression.

Methods

PD patients with two consecutive clinical visits while taking medication were recruited. Based on mean changes in Movement Disorder Society Unified Parkinson's Disease Rating Scale part 3 score and levodopa equivalent daily dosage, the participants were categorized into rapid progression, medium progression, and slow progression groups. Odors associated with the risk of PD motor progression were identified by calculating the odds ratios of UPSIT item identification between the rapid and slow progression groups. Receiver operating characteristic curve analysis of these odors was conducted to determine an optimal threshold for rapid motor progression.

Results

A total of 117 PD patients were screened for group classification. Preserved identification of neutral/pleasant odors including banana, peach, magnolia, and baby powder was significantly correlated with rapid motor progression. The risk of rapid progression increased with more detected risk odors. Detection of ≥1.5 risk odors could differentiate rapid progression from slow progression with a sensitivity of 85.7%, specificity of 45.8%, and area under the receiver operating characteristic curve of 0.687.

Conclusion

Preserved identification of neutral/pleasant odors may help to predict PD motor progression, and detection of ≥1.5 UPSIT motor progression risk odors could improve the predictive power. In PD patients with a similar level of motor disability during initial screening, preserved pleasant/neutral odor identification may imply relatively better cortical odor discriminative function, which may suggest the body-first (caudo-rostral) subtype with faster disease progression.

A machine learning approach to predict quality of life changes in patients with Parkinson's Disease

OBJECTIVE

Parkinson disease (PD) is a progressive neurodegenerative disorder with an annual incidence of approximately 0.1%. While primarily considered a motor disorder, increasing emphasis is being placed on its non-motor features. Both manifestations of the disease affect quality of life (QoL), which is captured in part II of the Unified Parkinson's Disease Rating Scale (UPDRS-II). While useful in the management of patients, it remains challenging to predict how QoL will change over time in PD. The goal of this work is to explore the feasibility of a machine learning algorithm to predict QoL changes in PD patients.

METHODS

In this retrospective cohort study, patients with at least 12 months of follow-up were identified from the Parkinson's Progression Markers Initiative database (N = 630) and divided into two groups: those with and without clinically significant worsening in UPDRS-II (n = 404 and n = 226, respectively). We developed an artificial neural network using only UPDRS-II scores, to predict whether a patient would clinically worsen or not at 12 months from follow-up.

RESULTS

Using UPDRS-II at baseline, at 2 months, and at 4 months, the algorithm achieved 90% specificity and 56% sensitivity.

INTERPRETATION

A learning model has the potential to rule in patients who may exhibit clinically significant worsening in QoL at 12 months. These patients may require further testing and increased focus.

Machine learning within the Parkinson's progression markers initiative: Review of the current state of affairs

The Parkinson's Progression Markers Initiative (PPMI) has collected more than a decade's worth of longitudinal and multi-modal data from patients, healthy controls, and at-risk individuals, including imaging, clinical, cognitive, and 'omics' biospecimens. Such a rich dataset presents unprecedented opportunities for biomarker discovery, patient subtyping, and prognostic prediction, but it also poses challenges that may require the development of novel methodological approaches to solve. In this review, we provide an overview of the application of machine learning methods to analyzing data from the PPMI cohort. We find that there is significant variability in the types of data, models, and validation procedures used across studies, and that much of what makes the PPMI data set unique (multi-modal and longitudinal observations) remains underutilized in most machine learning studies. We review each of these dimensions in detail and provide recommendations for future machine learning work using data from the PPMI cohort.

Motor Progression in Early-Stage Parkinson's Disease: A Clinical Prediction Model and the Role of Cerebrospinal Fluid Biomarkers

Background: The substantial heterogeneity of clinical symptoms and lack of reliable progression markers in Parkinson's disease (PD) present a major challenge in predicting accurate progression and prognoses. Increasing evidence indicates that each component of the neurovascular unit (NVU) and blood-brain barrier (BBB) disruption may take part in many neurodegenerative diseases. Since some portions of CSF are eliminated along the neurovascular unit and across the BBB, disturbing the pathways may result in changes of these substances.

Methods: Four hundred seventy-four participants from the Parkinson's Progression Markers Initiative (PPMI) study (NCT01141023) were included in the study. Thirty-six initial features, including general information, brief clinical characteristics and the current year's classical scale scores, were used to build five regression models to predict PD motor progression represented by the coming year's Unified Parkinson's Disease Rating Scale (MDS-UPDRS) Part III score after redundancy removal and recursive feature elimination (RFE)-based feature selection. Then, a threshold range was added to the predicted value for more convenient model application. Finally, we evaluated the CSF and blood biomarkers' influence on the disease progression model.

Results: Eight hundred forty-nine cases were included in the study. The adjusted R² values of three different categories of regression model, linear, Bayesian and ensemble, all reached 0.75. Models of the same category shared similar feature combinations. The common features selected among the categories were the MDS-UPDRS Part III score, Montreal Cognitive Assessment (MOCA) and Rapid Eye Movement Sleep Behavior Disorder Questionnaire (RBDSQ) score. It can be seen more intuitively that the model can achieve certain prediction effect through threshold range. Biomarkers had no significant impact on the progression model within the data in the study.

Conclusions: By using machine learning and routinely gathered assessments from the current year, we developed multiple dynamic models to predict the following year's motor progression in the early stage of PD. These methods will allow clinicians to tailor medical management to the individual and identify at-risk patients for future clinical trials examining disease-modifying therapies.

Machine Learning in Human Olfactory Research

The complexity of the human sense of smell is increasingly reflected in complex and high-dimensional data, which opens opportunities for data-driven approaches that complement hypothesis-driven research. Contemporary developments in computational and data science, with its currently most popular implementation as machine learning, facilitate complex data-driven research approaches. The use of machine learning in human olfactory research included major approaches comprising 1) the study of the physiology of pattern-based odor detection and recognition processes, 2) pattern recognition in olfactory phenotypes, 3) the development of complex disease biomarkers including olfactory features, 4) odor prediction from physico-chemical properties of volatile molecules, and 5) knowledge discovery in publicly available big databases. A limited set of unsupervised and supervised machine-learned methods has been used in these projects, however, the increasing use of contemporary methods of computational science is reflected in a growing number of reports employing machine learning for human olfactory research. This review provides key concepts of machine learning and summarizes current applications on human olfactory data.