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

Balancing the needs of data privacy and predictive utility is a centralchallenge for machine learning in healthcare. In particular, privacy concernshave led to a dearth of public datasets, complicated the construction ofmulti-hospital cohorts and limited the utilization of external machine learningresources. To remedy this, new methods are required to enable data owners, suchas hospitals, to share their datasets publicly, while preserving both patientprivacy and modeling utility. We propose NeuraCrypt, a private encoding schemebased on random deep neural networks. NeuraCrypt encodes raw patient data usinga randomly constructed neural network known only to the data-owner, andpublishes both the encoded data and associated labels publicly. From atheoretical perspective, we demonstrate that sampling from a sufficiently richfamily of encoding functions offers a well-defined and meaningful notion ofprivacy against a computationally unbounded adversary with full knowledge ofthe underlying data-distribution. We propose to approximate this family ofencoding functions through random deep neural networks. Empirically, wedemonstrate the robustness of our encoding to a suite of adversarial attacksand show that NeuraCrypt achieves competitive accuracy to non-private baselineson a variety of x-ray tasks. Moreover, we demonstrate that multiple hospitals,using independent private encoders, can collaborate to train improved x-raymodels. Finally, we release a challenge dataset to encourage the development ofnew attacks on NeuraCrypt.

NeuraCrypt: Hiding Private Health Data via Random Neural Networks for Public Training