HOSPICES CIVILS DE LYON

A dataset of 1,249 uveitis cases with unknown etiology was used, with 766 (61.3%) receiving an etiological diagnosis. The study employed the ULISSE screening protocol and SUN classification for uveitis. A neural network model, Multilayer Perceptron (MLP), was developed to predict uveitis etiologies, outperforming SVM and RF in accuracy. The study highlighted the risks of oversampling methods like SMOTE, which altered the dataset's real distribution, affecting diagnostic accuracy.

A French study links aromatic amines in hair dye to retinopathy, with three women developing multiple bilateral serous retinal detachments (SRDs) after use. The condition, RAHDAA, mimics MEK inhibitor–associated retinopathy and resolved after discontinuing the dyes. A 61-year-old woman's case showed SRDs resolving within 4 months, with no recurrence after switching to aromatic amine-free dye.

The study involves a 45-month trial with 8 centers, recruiting patients over 24 months for ziconotide or placebo treatment phases, each lasting 6 months, with a 1-month wash-out period. Patients undergo LP or continuous infusion tests to determine response, with responders implanted with continuous infusion pumps. The study includes detailed visit schedules, randomization procedures, and statistical analysis plans to evaluate pain reduction and patient satisfaction.