no code implementations • 12 Oct 2023 • Carlos Gómez-Rodríguez, Paul Williams
We evaluate a range of recent LLMs on English creative writing, a challenging and complex task that requires imagination, coherence, and style.
no code implementations • 28 Jun 2023 • Jordan Smith, Tom Naunton Morgan, Paul Williams, Qaiser Malik, Simon Rasalingham
AIM To analyse the performance of a deep-learning (DL) algorithm currently deployed as diagnostic decision support software in two NHS Trusts used to identify normal chest x-rays in active clinical pathways.
no code implementations • 31 Aug 2022 • Gaetan Dissez, Nicole Tay, Tom Dyer, Matthew Tam, Richard Dittrich, David Doyne, James Hoare, Jackson J. Pat, Stephanie Patterson, Amanda Stockham, Qaiser Malik, Tom Naunton Morgan, Paul Williams, Liliana Garcia-Mondragon, Jordan Smith, George Pearse, Simon Rasalingham
Design: This retrospective study evaluated the performance of 11 clinicians for detecting lung cancer from chest radiographs, with and without assistance from a commercially available AI algorithm (red dot, Behold. ai) that predicts suspected lung cancer from CXRs.
no code implementations • 31 Aug 2022 • Tom Dyer, Jordan Smith, Gaetan Dissez, Nicole Tay, Qaiser Malik, Tom Naunton Morgan, Paul Williams, Liliana Garcia-Mondragon, George Pearse, Simon Rasalingham
This study evaluates the robustness of an AI solution for the diagnosis of normal chest X-rays (CXRs) by comparing performance across multiple patient and environmental subgroups, as well as comparing AI errors with those made by human experts.