English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Rodent stroke models are important for evaluating treatments andunderstanding the pathophysiology and behavioral changes of brain ischemia, andmagnetic resonance imaging (MRI) is a valuable tool for measuring outcome inpreclinical studies. Brain extraction is an essential first step in mostneuroimaging pipelines; however, it can be challenging in the presence ofsevere pathology and when dataset quality is highly variable. Convolutionalneural networks (CNNs) can improve accuracy and reduce operator time,facilitating high throughput preclinical studies. As part of an ongoingpreclinical stroke imaging study, we developed a deep-learning mouse brainextraction tool by using a U-net CNN. While previous studies have evaluatedU-net architectures, we sought to evaluate their practical performance acrossdata types. We ask how performance is affected with data across: six imagingcenters, two time points after experimental stroke, and across four MRIcontrasts. We trained, validated, and tested a typical U-net model on 240multimodal MRI datasets including quantitative multi-echo T2 and apparentdiffusivity coefficient (ADC) maps, and performed qualitative evaluation with alarge preclinical stroke database (N=1,368). We describe the design anddevelopment of this system, and report our findings linking datacharacteristics to segmentation performance. We consistently found highaccuracy and ability of the U-net architecture to generalize performance in arange of 95-97% accuracy, with only modest reductions in performance based onlower fidelity imaging hardware and brain pathology. This work can help informthe design of future preclinical rodent imaging studies and improve theirscalability and reliability.

Evaluating U-net Brain Extraction for Multi-site and Longitudinal Preclinical Stroke Imaging