Dataset Information

Retinal texture biomarkers may help to discriminate between Alzheimer's, Parkinson's, and healthy controls.

ABSTRACT: A top priority in biomarker development for Alzheimer's disease (AD) and Parkinson's disease (PD) is the focus on early diagnosis, where the use of the retina is a promising avenue of research. We computed fundus images from optical coherence tomography (OCT) data and analysed the structural arrangement of the retinal tissue using texture metrics. We built clinical class classification models to distinguish between healthy controls (HC), AD, and PD, using machine learning (support vector machines). Median sensitivity is 88.7%, 79.5% and 77.8%, for HC, AD, and PD eyes, respectively. When the same subject has the same classification for both eyes, 94.4% (median) of the classifications are correct. A significant amount of information discriminating between multiple neurodegenerative states is conveyed by OCT imaging of the human retina, even when differences in thickness are not yet present. This technique may allow for simultaneously diagnosing Alzheimer's and Parkinson's diseases.

SUBMITTER: Nunes A

PROVIDER: S-EPMC6588252 | biostudies-literature | 2019

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Retinal texture biomarkers may help to discriminate between Alzheimer's, Parkinson's, and healthy controls.

Nunes Ana A Silva Gilberto G Duque Cristina C Januário Cristina C Santana Isabel I Ambrósio António Francisco AF Castelo-Branco Miguel M Bernardes Rui R

PloS one 20190621 6

A top priority in biomarker development for Alzheimer's disease (AD) and Parkinson's disease (PD) is the focus on early diagnosis, where the use of the retina is a promising avenue of research. We computed fundus images from optical coherence tomography (OCT) data and analysed the structural arrangement of the retinal tissue using texture metrics. We built clinical class classification models to distinguish between healthy controls (HC), AD, and PD, using machine learning (support vector machine ...[more]

PMID: 31226150

Similar Datasets

Project description:BackgroundPatients with Parkinson's disease (PD) and those with multiple system atrophy (MSA) show similar symptoms but have different clinical treatments. It will be helpful to discriminate between these two kinds of patients at an early or middle stage. The purpose of this study is to highlight the differences in posturographic characterization between patients with PD and those with MSA during quiet standing and perturbed standing.MethodsA total of clinically diagnosed 42 patients with PD and 32 patients with MSA participated in the experiment. Patients were asked to first stand on a static balance force platform and then on a dynamic balance (medial-lateral rocker) force platform to measure the center of pressure (COP) trajectory during an eyes-open (EO) state. The posturographic parameters were obtained under the two standing conditions for statistical analysis.ResultsFour posturographic variables were calculated and analyzed, namely, the standard deviation of COP position (SD), sway path of COP position (SP), an elliptical area covering the 95% COP position trajectory (EA), sway path of COP position (SP), and integral area of the power spectral density at 0-0.5 Hz frequency band (PSD). Except for variable EA, the other three variables are all in the medial-lateral (ML) direction. In the static balance experiment, there were no significant differences between the four variables between patients with PD and those with MSA. However, in the dynamic balance experiment, the obtained four variables all presented significant differences between patients with PD and those with MSA.ConclusionThe dynamic posturographic variables with significant differences between patients with PD and those with MSA imply that patients with MSA have worse postural control ability in the medial-lateral (ML) direction compared to patients with PD. The obtained dynamic indices may help supplemental clinical evaluation to discriminate between patients with MSA and those with PD.

Project description:This systematic review aims to provide a comprehensive understanding of the current literature regarding gut microbiota composition in patients with Parkinson's disease (PD) and Alzheimer's disease (AD) compared to healthy controls. To identify the relevant studies, a thorough search of PubMed, Medline, and Embase was conducted following the PRISMA guidelines. Out of 5627 articles, 73 studies were assessed for full-text eligibility, which led to the inclusion of 42 studies (26 PD and 16 AD studies). The risk of bias assessment showed a medium risk in 32 studies (20 PD studies and 12 AD studies), a low risk in 9 studies (5 PD studies and 4 AD studies), and 1 PD study with a high risk. Among the PD studies, 22 out of 26 studies reported a different gut microbiota composition between the PD cases and the healthy controls, and 15 out of 16 AD studies reported differences in gut microbiota composition between the AD cases and the healthy controls. The PD and AD studies consistently identified the phyla Bacteroidetes, Firmicutes, and Proteobacteria as prevalent in the gut microbiota in both the healthy groups and the case groups. Microbial dysbiosis was specifically characterized in the PD studies by a high abundance of Akkermansia, Verrucomicrobiaceae, Lachnospiraceae, and Ruminococcaceae in the cases and a high abundance of Blautia, Coprococcus, Prevotellaceae, and Roseburia in the controls. Similarly, Bacteroides and Acidobacteriota were abundant in the AD cases, and Acidaminococcaceae, Firmicutes, Lachnospiraceae, and Ruminiclostridium were abundant in the AD controls. The microbial signature assessment showed the association of several microbial taxa, including Akkermansia, Lachnospiraceae, Verrucomicrobiaceae, Bifidobacterium, Ruminococcacea, and Verrucomicrobia with PD and Ruminococcaceae, Bacteroides, and Actinobacteria with AD. The microbial diversity evaluations in the PD and AD studies indicated comparable alpha diversity in some groups and distinct gut microbiota composition in others, with consistent beta diversity differences between the cases and the controls across multiple studies. The bacterial signatures identified in this study that are associated with PD and AD may offer promising prospects for efficient management and treatment approaches.

Dataset Information

Retinal texture biomarkers may help to discriminate between Alzheimer's, Parkinson's, and healthy controls.

Publications

Retinal texture biomarkers may help to discriminate between Alzheimer's, Parkinson's, and healthy controls.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets