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

Intra-operative ultrasound is an increasingly important imaging modality inneurosurgery. However, manual interaction with imaging data during theprocedures, for example to select landmarks or perform segmentation, isdifficult and can be time consuming. Yet, as registration to other imagingmodalities is required in most cases, some annotation is necessary. We proposea segmentation method based on DeepVNet and specifically evaluate theintegration of pre-training with simulated ultrasound sweeps to improveautomatic segmentation and enable a fully automatic initialization ofregistration. In this view, we show that despite training on coarse andincomplete semi-automatic annotations, our approach is able to capture thedesired superficial structures such as \textit{sulci}, the \textit{cerebellartentorium}, and the \textit{falx cerebri}. We perform a five-foldcross-validation on the publicly available RESECT dataset. Trained on thedataset alone, we report a Dice and Jaccard coefficient of $0.45 \pm 0.09$ and$0.30 \pm 0.07$ respectively, as well as an average distance of $0.78 \pm0.36~mm$. With the suggested pre-training, we computed a Dice and Jaccardcoefficient of $0.47 \pm 0.10$ and $0.31 \pm 0.08$, and an average distance of$0.71 \pm 0.38~mm$. The qualitative evaluation suggest that with pre-trainingthe network can learn to generalize better and provide refined and morecomplete segmentations in comparison to incomplete annotations provided asinput.

Fully Automatic Segmentation of 3D Brain Ultrasound: Learning from Coarse Annotations