Observed arrival · 2026-09-15
A browser-based second look at leukemia cells
A neural network ranks individual blood-cell images by their resemblance to leukemic lymphoblasts, putting the cells most worth reviewing first.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
Field notes
The model takes an isolated cell image and produces a 0–1 resemblance score, which can be used to reorder a smear for human review. The page reports results on 1,882 cells from 11 patients excluded from training, including 851 correctly flagged leukemic cells and 36 normal cells flagged in error at a 0.770 threshold. Its 47.7 MB network runs in the browser, with the site stating that images are neither uploaded nor sent to a server.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use
One card from the complete issue