A groundbreaking study led by a team of researchers at the University of Oxford has unveiled a new method for accurately tracking the progression of Parkinson’s Disease. Using specially trained machine learning algorithms to analyze data collected from sensor devices worn by patients, this research, led by Professor Chrystalina Antoniades in Oxford’s Nuffield Department of Clinical Neurosciences, introduces innovative techniques that can complement traditional clinical rating scales. These tools not only enhance the precision of diagnosis but also enable the tracking of disease progression in individuals with Parkinson’s Disease.
Monitoring the progression of motor symptoms in individuals with neurological disorders like Parkinson’s Disease serves two crucial purposes: it empowers clinicians to confidently assess how the disease is advancing in individual patients, and it enables researchers conducting clinical trials to gauge the effectiveness of therapeutic interventions.
EARLY IDENTIFICATION OF EFFECTIVE TREATMENT
This development carries significant implications for clinical trials, where bringing a new drug for Parkinson’s Disease from the laboratory to clinical use involves years of effort and extensive resources. Early identification of effective treatments is critical to accelerating their development. Professor Chrystalina Antoniades expressed optimism about the new objective measuring tools, stating, “I hope this will be made easier with these new objective measuring tools.”
Currently, clinicians rely on rating scales, which are scoring systems based on a physical examination, to evaluate the key symptoms in people with Parkinson’s Disease. However, these scales have limitations, including subjectivity in assessment and uneven spacing of scores. These factors can lead to delayed detection of disease progression and restrictions on statistical analysis in clinical trials.
THE STUDY
Professor Antoniades and her NeuroMetrology Lab have conducted experiments using sensor devices placed on the trunk, wrists, and feet of patients, combined with machine learning, to assess whether these tools can track motor symptom progression more accurately than traditional rating scales.
Previous studies have shown that wearable devices, combined with machine learning, can aid in accurate diagnosis. These techniques can distinguish between healthy older adults, individuals with varying degrees of Parkinson’s Disease severity, and those with Parkinsonian-like disorders.
In this new study, the researchers examined whether data collected during walking and standing tasks could not only aid in diagnosis but also track the progression of motor symptoms in Parkinson’s Disease over time. Study participants underwent intensive assessments with clinic visits every three months to determine the shortest timeframe for detecting disease progression. The research demonstrated that their technique could detect progression in as little as 15 months.
This study is part of the Oxford Quantification in Parkinsonism Study (OxQUIP), a longitudinal research project funded by UCB.
