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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Deep learning has shown remarkable results for image analysis and is expectedto aid individual treatment decisions in health care. To achieve this, deeplearning methods need to be promoted from the level of mere associations tobeing able to answer causal questions. We present a scenario with real-worldmedical images (CT-scans of lung cancers) and simulated outcome data. Throughthe sampling scheme, the images contain two distinct factors of variation thatrepresent a collider and a prognostic factor. We show that when this collidercan be quantified, unbiased individual prognosis predictions are attainablewith deep learning. This is achieved by (1) setting a dual task for the networkto predict both the outcome and the collider and (2) enforcing independence ofthe activation distributions of the last layer with ordinary least squares. Ourmethod provides an example of combining deep learning and structural causalmodels for unbiased individual prognosis predictions.

Controlling for Biasing Signals in Images for Prognostic Models: Survival Predictions for Lung Cancer with Deep Learning