Finally got around to watching the update on the KETO-CTA cohort. Not much to say for now – I’ll take a closer look once their paper is published.
Thoughts:
1⃣ Why are they focusing on relative measures, especially when several of the authors have agreed that this is misleading?
2⃣ From what I tell, the absolute change that picked up by Cleerly in those with a CAC score of 0 = minimal. So the reported >100% relative change feels quite misleading in that context.
3⃣ I tend to think Cleerly may be a more sensitive software package – that it picks up small absolute changes, which, of course, tend to translate into large relative changes.
4⃣ I'm not sure it makes sense to compare one AI tool to another – without an internal control, the best we can do, I imagine, is compare Cleerly data to other cohorts that also used Cleerly. Likewise, the same should be done for the HeartFlow and QAngio analyses.
5⃣ Would any of these re-analyses be happening if the Cleerly results had aligned with the ideological bias of the authors – and the study's funders? I don’t think it’s overly cynical to suspect not.