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

Today, cataract surgery is the most frequently performed ophthalmic surgeryin the world. The cataract, a developing opacity of the human eye lens,constitutes the world's most frequent cause for blindness. During surgery, thelens is removed and replaced by an artificial intraocular lens (IOL). Toprevent patients from needing strong visual aids after surgery, a preciseprediction of the optical properties of the inserted IOL is crucial. There hasbeen lots of activity towards developing methods to predict these propertiesfrom biometric eye data obtained by OCT devices, recently also by employingmachine learning. They consider either only biometric data or physical models,but rarely both, and often neglect the IOL geometry. In this work, we proposeOpticNet, a novel optical refraction network, loss function, and trainingscheme which is unsupervised, domain-specific, and physically motivated. Wederive a precise light propagation eye model using single-ray raytracing andformulate a differentiable loss function that back-propagates physicalgradients into the network. Further, we propose a new transfer learningprocedure, which allows unsupervised training on the physical model andfine-tuning of the network on a cohort of real IOL patient cases. We show thatour network is not only superior to systems trained with standard proceduresbut also that our method outperforms the current state of the art in IOLcalculation when compared on two biometric data sets.

Domain-specific loss design for unsupervised physical training: A new approach to modeling medical ML solutions