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 Plus annual

Presents the access of premium content as premium feature

Premium Content

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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

A commonly adopted approach to carry out detection tasks in medical imagingis to rely on an initial segmentation. However, this approach strongly dependson voxel-wise annotations which are repetitive and time-consuming to draw formedical experts. An interesting alternative to voxel-wise masks are so-called"weak" labels: these can either be coarse or oversized annotations that areless precise, but noticeably faster to create. In this work, we address thetask of brain aneurysm detection as a patch-wise binary classification withweak labels, in contrast to related studies that rather use supervisedsegmentation methods and voxel-wise delineations. Our approach comes with thenon-trivial challenge of the data set creation: as for most focal diseases,anomalous patches (with aneurysm) are outnumbered by those showing no anomaly,and the two classes usually have different spatial distributions. To tacklethis frequent scenario of inherently imbalanced, spatially skewed data sets, wepropose a novel, anatomically-driven approach by using a multi-scale andmulti-input 3D Convolutional Neural Network (CNN). We apply our model to 214subjects (83 patients, 131 controls) who underwent Time-Of-Flight MagneticResonance Angiography (TOF-MRA) and presented a total of 111 unrupturedcerebral aneurysms. We compare two strategies for negative patch sampling thathave an increasing level of difficulty for the network and we show how thischoice can strongly affect the results. To assess whether the added spatialinformation helps improving performances, we compare our anatomically-informedCNN with a baseline, spatially-agnostic CNN. When considering the morerealistic and challenging scenario including vessel-like negative patches, theformer model attains the highest classification results (accuracy$\simeq$95\%,AUROC$\simeq$0.95, AUPR$\simeq$0.71), thus outperforming the baseline.

An anatomically-informed 3D CNN for brain aneurysm classification with weak labels