2025‐08‐14 - aasiyahrashan/PhD-Aasiyah GitHub Wiki

Meeting with Steve, Abi, Matt

• Spoke about the ventilation prediction project
• We are having issues with justifying why we're using both datasets, and what exactly we want to find out
• Possible alternative analyses include Matt's suggestion about survival analysis with competing risks, Steve's suggestion to use day 4 data instead of day 1, and the possibility of comparing the prediction model to a simple rule based method used in trials.
• Agreed that we should flip the order we do this, and work on the target trial first. Then try to build the story.
• But I should email back in parallel and ask someone to help me put the story for the prediction project together.