Following this puts new work involving it at the top of your briefing, with a note saying why it is there. Links are taken from the source record, never inferred.
A well-designed, large single-cohort discovery study with independent validation that identifies clinically relevant body composition phenotypes and their cardiometabolic risk associations, but uses surrogate endpoints and spatial clustering metrics rather than hard clinical outcomes.
This is a machine learning systems paper presenting a novel architecture with computational experiments on historical financial data, not a clinical trial or validated trading study with prospective performance or real-world deployment evidence.