Elemental fingerprint: Reassessment of a cerebrospinal fluid biomarker for Parkinson's disease

Maass, Fabian; Michalke, Bernhard; Willkommen, Desiree; Leha, Andreas; Schulte, Claudia; Tönges, Lars; Mollenhauer, Brit; Trenkwalder, Claudia; Rückamp, Daniel; Börger, M. Elisa; Zerr, Inga; Bähr, Mathias; Lingor, Paul

doi:10.1016/j.nbd.2019.104677

Cited by 31 publications

(20 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…All 5 studies used AUC, instead of accuracy, to evaluate machine learning models ( Table 7 ). The average AUC was 0.8 (0.1), the lowest AUC was 0.6825 (Maass et al, 2020 ) and the highest AUC was 0.839 (Maass et al, 2018 ), respectively. Two studies obtained the highest per-study AUC with ensemble learning, 2 studies with SVM and 1 study with regression ( Figure 4B ).…”

Section: Resultsmentioning

confidence: 98%

“…As its name implies, machine learning allows for a computer program to learn and extract meaningful representation from data in a semi-automatic manner. For the diagnosis of PD, machine learning models have been applied to a multitude of data modalities, including handwritten patterns (Drotár et al, 2015 ; Pereira et al, 2018 ), movement (Yang et al, 2009 ; Wahid et al, 2015 ; Pham and Yan, 2018 ), neuroimaging (Cherubini et al, 2014a ; Choi et al, 2017 ; Segovia et al, 2019 ), voice (Sakar et al, 2013 ; Ma et al, 2014 ), cerebrospinal fluid (CSF) (Lewitt et al, 2013 ; Maass et al, 2020 ), cardiac scintigraphy (Nuvoli et al, 2019 ), serum (Váradi et al, 2019 ), and optical coherence tomography (OCT) (Nunes et al, 2019 ). Machine learning also allows for combining different modalities, such as magnetic resonance imaging (MRI) and single-photon emission computed tomography (SPECT) data (Cherubini et al, 2014b ; Wang et al, 2017 ), in the diagnosis of PD.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Machine Learning for the Diagnosis of Parkinson's Disease: A Review of Literature

Mei

Desrosiers

Frasnelli

2021

Front. Aging Neurosci.

248

View full text Add to dashboard Cite

Diagnosis of Parkinson's disease (PD) is commonly based on medical observations and assessment of clinical signs, including the characterization of a variety of motor symptoms. However, traditional diagnostic approaches may suffer from subjectivity as they rely on the evaluation of movements that are sometimes subtle to human eyes and therefore difficult to classify, leading to possible misclassification. In the meantime, early non-motor symptoms of PD may be mild and can be caused by many other conditions. Therefore, these symptoms are often overlooked, making diagnosis of PD at an early stage challenging. To address these difficulties and to refine the diagnosis and assessment procedures of PD, machine learning methods have been implemented for the classification of PD and healthy controls or patients with similar clinical presentations (e.g., movement disorders or other Parkinsonian syndromes). To provide a comprehensive overview of data modalities and machine learning methods that have been used in the diagnosis and differential diagnosis of PD, in this study, we conducted a literature review of studies published until February 14, 2020, using the PubMed and IEEE Xplore databases. A total of 209 studies were included, extracted for relevant information and presented in this review, with an investigation of their aims, sources of data, types of data, machine learning methods and associated outcomes. These studies demonstrate a high potential for adaptation of machine learning methods and novel biomarkers in clinical decision making, leading to increasingly systematic, informed diagnosis of PD.

show abstract

Section: Resultsmentioning

confidence: 98%

Section: Introductionmentioning

confidence: 99%

Machine Learning for the Diagnosis of Parkinson's Disease: A Review of Literature

Mei

Desrosiers

Frasnelli

2021

Front. Aging Neurosci.

248

View full text Add to dashboard Cite

show abstract

“…We performed a screening of the abstracts of all original articles, and publications in abstract form were excluded from further analysis. The initial selection included 76 studies measuring CSF, serum/plasma or urine levels of iron, ferritin, transferrin, lactoferrin, haptoglobin, and or hepcidine [6,9–83]. Table 2 summarizes the methods used for the determinations, the main results, and observations in some of these studies, including the reason for exclusion of several of these studies from the meta‐analysis.…”

Section: Methodsmentioning

confidence: 99%

Biological fluid levels of iron and iron‐related proteins in Parkinson’s disease: Review and meta‐analysis

Jiménez‐Jiménez

Alonso‐Navarro

García‐Martín

et al. 2020

Euro J of Neurology

View full text Add to dashboard Cite

Background and purpose Several studies suggested a role or iron in the pathogenesis or Parkinson's disease (PD), and substantia nigra iron concentrarions have been found increased in PD. However, the results on cerebrospinal (CSF) and serum/plasma iron levels in PD patients have been controversial. The aim of this systematic review and meta‐analysis was to establish the CSF and serum/plasma levels of iron and iron‐related proteins (ferritin, transferrin, lactoferrin, haptoglobin, and hepcidine) levels, and the urine levels of iron, in patients with PD. Methods Four databases (PubMed, EMBASE, MedLine, and Web of Science – Core Collection) were reviewed for studies published from 1966 to October 5, 2020. References of interest were identified. A meta‐analysis of eligible studies was performed according to the Preferred Reporting Items for Systematic Reviews and Meta‐analyses (PRISMA) and Meta‐analysis of Observational Studies in Epidemiology (MOOSE) guidelines, using the R software package meta. Results A non‐significant trend towards higher CSF iron levels and marginally significantly lower serum/plasma iron levels was observed in patients with PD compared with age‐ and sex‐matched controls. CSF and serum/plasma ferritin and transferrin concentrations, and serum/plasma lactoferrin and haptoglobin concentrations did not differ significantly between PD patients and controls. Conclusion The findings of this study suggest an association between decreased serum/plasma iron levels and, possibly, higher CSF iron levels with risk of PD.

show abstract

“…The main trend in the development of an early and differential diagnosis of PD is the search for biomarkers as changes in the level of certain substances in the body fluids, mainly in the blood and cerebrospinal fluid. However, data published by various authors are often contradictory [ 14 , 17 , 38 , 39 , 40 , 41 , 42 , 43 , 46 , 47 , 48 ]. This is probably due to the fact that the concentration of analytes in the blood and, to a lesser extent, in the cerebrospinal fluid is an integral index of a wide range of metabolic processes associated with degeneration and plasticity of central and peripheral neurons, as well as desympathization of the internal organs [ 8 , 20 ].…”

Section: Discussionmentioning

confidence: 99%

“…Although there are still no studies of catecholamines in TF in PD patients, numerous, though conflicting data were accumulated on a change in the level of catecholamines in the cerebrospinal fluid and blood in PD patients ( n = 9–53), which are considered as potential diagnostic markers [ 38 , 39 , 40 , 41 , 42 , 43 ]. Based on the above, the next objective of this study was to definitely determine in the enlarged cohort of PD patients whether changes in catecholamine levels in the TF are characteristic of PD and could be considered as potential diagnostic biomarkers.…”

Section: Introductionmentioning

confidence: 99%

A Pilot Study of Changes in the Level of Catecholamines and the Activity of α-2-Macroglobulin in the Tear Fluid of Patients with Parkinson’s Disease and Parkinsonian Mice

Bogdanov

Kim

Nodel

et al. 2021

IJMS

View full text Add to dashboard Cite

Development of differential and early (preclinical) diagnostics of Parkinson’s disease (PD) is among the priorities in neuroscience. We searched for changes in the level of catecholamines and α-2-macroglobulin activity in the tear fluid (TF) in PD patients at an early clinical stage. It was shown that TF in patients is characterized by an increased level of noradrenaline mainly on the ipsilateral side of pronounced motor symptoms (72%, p = 0.049), a decreased level of adrenaline on both sides (ipsilateral—53%, p = 0.004; contralateral—42%, p = 0.02), and an increased α-2-macroglobulin activity on both sides (ipsilateral—53%, p = 0.03; contralateral—56%, p = 0.037) compared to controls. These changes are considered as potential biomarkers for differential diagnosis. Similar changes in the TF were found in 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-treated mice when modeling clinical and preclinical stages of PD. These data show the adequacy of models to the pathogenesis of PD along the selected metabolic pathways, and also suggest that the found TF changes can be considered as potential biomarkers for preclinical diagnosis of PD. In Parkinsonian mice, the level of catecholamines also changes in the lacrimal glands, which makes it possible to consider them as one of the sources of catecholamines in the TF.

show abstract

Elemental fingerprint: Reassessment of a cerebrospinal fluid biomarker for Parkinson's disease

Cited by 31 publications

References 45 publications

Machine Learning for the Diagnosis of Parkinson's Disease: A Review of Literature

Machine Learning for the Diagnosis of Parkinson's Disease: A Review of Literature

Biological fluid levels of iron and iron‐related proteins in Parkinson’s disease: Review and meta‐analysis

A Pilot Study of Changes in the Level of Catecholamines and the Activity of α-2-Macroglobulin in the Tear Fluid of Patients with Parkinson’s Disease and Parkinsonian Mice

Contact Info

Product

Resources

About