The headline is broadly defensible, but the qualifications matter. Effect sizes vary by population, the strongest claims rest on shorter trials, and credible voices push back on how it's typically framed.
if you take prostate cancers and you do AI based hierarchical clustering just looking for patterns do you see different types of cancer not you know grade or stage but different types from the perspective of the transcriptome of prostate cancer and you do so you see two general themes one you see luminal like prostate Cancers and basil like prostate Cancers and then within that subclassification of luminal Basil there are kind of effectively aggressive luminals
Every Sunday: the week’s new conflicts and verdict changes — and nothing else.
Native comments, Twitter mentions, and Reddit threads about this claim — surfaced together so the conversation isn't fragmented across platforms.
Bookmarking — the dossier-vs-overview split is the right call. Most of the time I want overview; sometimes I want receipts.
Would love a "what would change this verdict" RSS feed. Sign me up if it exists.
if you take prostate cancers and you do aib based hierarchical clustering just looking for patterns do you see different types of cancer not you know grade or stage but different types from the perspective of the transcriptome of prostate cancer and you do so you see two general themes one you see luminal like prostate Cancers and basil like prostate cancers and then within that subclassification of luminal Basil there are kind of effectively aggressive luminals
Whole-body MRI screening in healthy adults produces more incidentaloma harm than cancer-mortality benefit.
Starting colonoscopy screening at 45 (vs 50) prevents enough early-onset cancers to justify the population cost.
Multi-cancer liquid-biopsy tests like Galleri detect early cancers at a stage that meaningfully improves survival.