Umetrics Suite Blog

Data analytics improves differentiation between ALS and Parkinson’s, leading to earlier diagnosis

December 5, 2018

Could data analytics aid in the diagnosis of severe neurological diseases? In a recent study, a research group at Umeå University has conducted statistical data analysis of biomarkers from patients suffering from Amyotrophic lateral sclerosis (ALS, also known as Lou Gehrig’s disease) and Parkinson’s disease to investigate whether data analytics could help in the diagnosis of – and help distinguish between – the two diseases.

parkinsons

A recent Swedish study shows that data analytics can help with earlier diagnosis of patients with Parkinson's Disease and ALS.

Early diagnosis could improve the quality of life for patients

Early diagnosis can not only prolong a patient’s life, it can also lead to a better quality of life if the patient receives the right treatment as early as possible. This is true not least for neurodegenerative disorders where early treatment can postpone disease symptoms that make it hard for the patient to lead a normal life.

Discovering new methods for differentiating Parkinson’s from ALS

Amyotrophic lateral sclerosis (ALS) and Parkinson’s disease (PD) are both degenerative disorders with no known cause. ALS and PD are difficult to diagnose and to distinguish between each other and there is therefore a need for new diagnostic methods that can help differentiate the diseases. Recently, a research group at the Umeå University, Sweden, has investigated whether statistical data analysis of metabolomes could aid in the diagnosis – and differential diagnosis – of ALS and PD.

Data analytics reveals hidden patterns

The research group analyzed mass spectrometry data of metabolomes from the cerebrospinal fluid (CSF) and plasma from patients with ALS and PD as well as from a control group from samples collected on a regular basis at the Umeå University Hospital. For the statistical data analysis, a statistical method called Orthogonal Partial Least Squares (OPLS) was used. The benefit of using OPLS for statistical data analysis is that it is good at filtering away non-correlated information and finding hidden patterns in the data.

Identification of metabolomes that separated the patient groups

The research group found 144 unique metabolites in CSF and 188 in plasma. They also found certain metabolites that separated ALS and PD patients from their matched controls, but of particular interest was the metabolomes in CSF and plasma that differentiated matched ALS and PD patients. The OPLS models found 12 identified metabolites that were significantly different in plasma for the two diseases and 6 in the CSF. Although none of the statistical data analysis models could completely separate ALS and PD patients from their matched controls, the study showed, similar to previous studies, that perturbed metabolomes could be used for the diagnosis of ALS and PD.

Learn more

Read the case story for more insights on the study.

Get the Case Story

 

References

  1. Wuolikainen et al., Multi-platform mass spectrometry analysis of the CSF and plasma metabolomes of rigorously matched amyotrophic lateral sclerosis, Parkinson's disease and control subjects. Mol Biosyst. 2016 Apr;12(4):1287-98

 

Topics: SIMCA, Data Interpretation & Analysis, Data Analytics

Marie Wensley

Written by Marie Wensley

Marketing Manager at Sartorius Stedim Data Analytics

Leave a comment