BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Deep Learning-Based Virtual Histopathology Staining of Stimulated 
 Raman Scattering Microscopy - Elisa Marion Billard\, epfl
DTSTART:20260407T160000Z
DTEND:20260407T163000Z
UID:TALK246388@talks.cam.ac.uk
CONTACT:Pietro Lio
DESCRIPTION:Traditional cancer diagnostics depend on time-consuming chemic
 al staining that can delay critical surgical decisions. As part of the EU-
 funded CHARM project\, this talk introduces a "virtual staining" pipeline 
 that uses 38-channel broadband Stimulated Raman Scattering (SRS) microscop
 y to generate label-free histological images. By implementing a high-preci
 sion registration framework\, we transitioned from unsupervised style tran
 sfer to a supervised deep learning approach (U-Net and CycleGAN). Our resu
 lts show that while standard models provide high structural accuracy\, hyb
 rid adversarial models produce superior visual clarity and sharper cellula
 r details. We further evaluate these virtual stains using automated nuclei
  segmentation (Cellpose)\, revealing a key divergence between human and al
 gorithmic interpretability. This research demonstrates a chemically-ground
 ed path for rapid\, non-destructive morphological feedback in intraoperati
 ve settings\, bypassing the logistical bottlenecks of conventional patholo
 gy.
LOCATION:LT1
END:VEVENT
END:VCALENDAR