Objective: To evaluate whether primary tumor treatment provides a survival benefit in uveal melanoma by comparing patients who declined treatment (Natural History Study, NHS) with those who received treatment in the Collaborative Ocular Melanoma Study (COMS) for medium-sized choroidal melanomas. Methods: A retrospective cohort study was conducted using reverse-engineered individual-level survival data from cumulative all-cause mortality curves in the original COMS and NHS publications. Censoring patterns were estimated from numbers at risk and descriptive statistics. To ensure a conservative approach, a Bonferroni-corrected significance level of 0.017 was applied. Further, if the Cox regression hazard ratio for all-cause mortality, the unadjusted risk ratio for death at 8 years, and the 95% confidence intervals (CIs) of the Kaplan-Meier curves did not indicate a smaller survival difference for the resulting data than reported in the original studies, we iteratively adjusted the NHS cohort data by reducing the 8-year cumulative mortality by one percentage point until these criteria were met. Results: Kaplan-Meier analysis revealed significantly higher cumulative mortality in the NHS cohort compared to the COMS cohort (Log-rank P = 0.012). When restricting the analysis to the first 8 years to account for unclear censoring patterns beyond this period, the NHS cohort still demonstrated worse survival (P = 0.008). Sensitivity analysis varying censoring times by ±25% over 1,000 iterations confirmed worse survival in the NHS cohort in 100% of cases. Conclusions: Patients who declined primary tumor treatment had significantly worse survival than those who received treatment. These findings support the hypothesis that primary tumor treatment confers a survival benefit in uveal melanoma.

Meaning understood in terms of teachability and learnability is crucial to Wittgenstein’s later work. As regards the resolution of philosophical problems – and epistemological problems in particular - this approach seems to posit a hierarchy of meaning that excludes endless deferral. This is the basis of Wittgenstein’s attack on philosophical scepticism. Derrida’s approach to language seems to require both non-hierarchy and endless deferral. Consequently fundamental to his concept of origin is identity and difference simultaneously, irreducibly, non-simply. One question is whether it is possible for there to be a compromise between the two philosophers: a hierarchy of meaning that does not in principle exclude endless deferral.

Wearable technology will become available and allow prolonged electroencephalography (EEG) monitoring in the home environment of patients with epilepsy. Neurologists analyse the EEG visually and annotate all seizures, which patients often under report. Visual analysis of a 24 hour EEG recording typically takes one to two hours. Reliable automated seizure detection algorithms will be crucial to reduce this analysis. We study a dataset of behind-the-ear EEG measurements. Our first aim was to develop a methodology to reduce the EEG dataset by classifying part of the data automatically, while retaining 100% detection sensitivity (DS). Prediction confidences are determined by temperature scaling of the classification model outputs and trust scores. A DS of approximately 90% (99%) can be achieved when automatically classifying around 90% (60%) of the data. Perfect DS can be achieved when automatically classifying 50% of the data. Our second contribution demonstrates that a common modelling strategy, where predictions from several short EEG segments are used to obtain a final prediction, can be improved by filtering out untrustworthy segments with low trust scores. The false detection rate shows a relative decrease between 21% and 43%, and the DS shows a small increase or decrease.