Biology subjects

Oeltze-Jafra, S.

Publications and source records attributed to Oeltze-Jafra, S..

3 recordsLinked to original sources

bioRxiv2024

Fully Automated MRI-based Analysis of the Locus Coeruleus in Aging and Alzheimer's Disease Dementia using ELSI-Net

INTRODUCTIONThe Locus Coeruleus (LC) is linked to the development and pathophysiology of neurodegenerative diseases such as Alzheimers Disease (AD). Magnetic Resonance Imaging based LC features have shown potential to assess LC integrity in vivo. METHODSWe present a Deep Learning based LC segmentation and feature extraction method: ELSI-Net and apply it to healthy aging and AD dementia datasets. Agreement to expert raters and previously published LC atlases were assessed. We aimed to reproduce previously reported differences in LC integrity in aging and AD dementia and correlate extracted features to cerebrospinal fluid (CSF) biomarkers of AD pathology. RESULTSELSI-Net demonstrated high agreement to expert raters and published atlases. Previously reported group differences in LC integrity were detected and correlations to CSF biomarkers were found. DISCUSSIONAlthough we found excellent performance, further evaluations on more diverse datasets from clinical cohorts are required for a conclusive assessment of ELSI-Nets general applicability. HighlightsO_LIthorough evaluation of a fully automatic LC segmentation method termed ELSI-Net in aging and AD dementia C_LIO_LIELSI-Net outperforms previous work and shows high agreement with manual ratings and previously published LC atlases C_LIO_LIELSI-Net replicates previously shown LC group differences in aging and AD C_LIO_LIELSI-Nets LC volume correlates with CSF biomarkers of AD pathology C_LI RESEARCH IN CONTEXTO_LISystematic Review: The authors reviewed the literature using traditional sources (e.g. Pubmed, Google Scholar). Although there are several publications introducing semi-automatic methods for LC segmentation, the application of Deep Learning methods is underexplored. To the best of our knowledge, this is the first paper using a Deep Learning based approach for automated LC segmentation in AD dementia. C_LIO_LIInterpretation: Our work introduces and evaluates an improved automatic, Deep Learning based LC segmentation and analysis approach. The results suggest a very high potential for practical applicability, e.g. in large-scale clinical studies for neurodegenerative diseases. C_LIO_LIFuture Directions: ELSI-Net can be used to assess LC integrity on large- or small-scale studies in Alzheimers Disease dementia. To ensure robust performance, ELSI-Net should be further evaluated in larger, more diverse datasets comprising varying LC MRI protocols and clinical populations. C_LI

bioinformatics↗

bioRxiv2022

Overcoming the Reproducibility Crisis - Results of the first Community Survey of the German National Research Data Infrastructure for Neuroscience.

The lack of reproducibility of research results is a serious problem - known as "the reproducibility crisis". The German National Research Data Infrastructure (NFDI) initiative implemented by the German Research Foundation (DFG) aims to help overcoming this crisis by developing sustainable solutions for research data management (RDM). NFDI comprises domain specific consortia across all science disciplines. In the field of neuroscience, NFDI Neuroscience (NFDI-Neuro) contributes to the strengthening of systematic and standardized RDM in its research communities. NFDI-Neuro conducted a comprehensive survey amongst the neuroscience community to determine the current needs, challenges, and opinions with respect to RDM. The outcomes of this survey are presented here. The German neuroscience community perceives barriers with respect to RDM and data sharing mainly linked to (1) lack of data and metadata standards, (2) lack of community adopted provenance tracking methods, 3) lack of a privacy preserving research infrastructure for sensitive data (4) lack of RDM literacy and (5) lack of required time and resources for proper RDM. NFDI-Neuro aims to systematically address these barriers by leading and contributing to the development of standards, tools, and infrastructure and by providing training, education, and support, as well as additional resources for RDM to its research community. The RDM work of NFDI-Neuro is conducted in close collaboration with its partner EBRAINS AISBL, the coordinating entity of the EU Flagship Human Brain Project, and its Research Infrastructure (RI) EBRAINS with more than 5000 registered users and developers from more than 70 countries of all continents. While NFDI-Neuro aims to address the German national needs, it closely aligns with the international community and the topics of the Digital Europe Program and EU Data Spaces. Significance StatementA comprehensive survey amongst the neuroscience community in Germany determined the current needs, challenges, and opinions with respect to standardized research data management (RDM) to overcome the reproducibility crisis. Significant deficits were pointed out concerning the perceived lack of standards for data and metadata, lack of provenance tracking and versioning of data, lack of protected digital research infrastructure for sensitive data and the lack of education and resources for proper RDM. Yet, at the same time, an overwhelming majority of community members indicated that they would be willing to share their data with other researchers and are interested to increase their RDM skills. Thus, the survey results suggest that training, the provision of standards, tools, infrastructure and additional resources for RDM holds the potential to significantly facilitate reproducible research in neuroscience.

neuroscience↗

bioRxiv2019

Automated Quantification of Enlarged Perivascular Spaces in Clinical Brain MRI across Sites

Enlarged perivascular spaces (PVS) are structural brain changes visible in MRI, and are a marker of cerebral small vessel disease. Most studies use time-consuming and subjective visual scoring to assess these structures. Recently, automated methods to quantify enlarged perivascular spaces have been proposed. Most of these methods have been evaluated only in high resolution scans acquired in controlled research settings. We evaluate and compare two recently published automated methods for the quantification of enlarged perivascular spaces in 76 clinical scans acquired from 9 different scanners. Both methods are neural networks trained on high resolution research scans and are applied without fine-tuning the networks parameters. By adapting the preprocessing of clinical scans, regions of interest similar to those computed from research scans can be processed. The first method estimates only the number of PVS, while the second method estimates simultaneously also a high resolution attention map that can be used to detect and segment PVS. The Pearson correlations between visual and automated scores of enlarged perivascular spaces were higher with the second method. With this method, in the centrum semiovale, the correlation was similar to the inter-rater agreement, and also similar to the performance in high resolution research scans. Results were slightly lower than the inter-rater agreement for the hippocampi, and noticeably lower in the basal ganglia. By computing attention maps, we show that the neural networks focus on the enlarged perivascular spaces. Assessing the burden of said structures in the centrum semiovale with the automated scores reached a satisfying performance, could be implemented in the clinic and, e.g., help predict the bleeding risk related to cerebral amyloid angiopathy.

neuroscience↗