An automated tissue-to-diagnosis pipeline using intraoperative stimulated Raman histology and deep learning.
Stimulated Raman histology
convolutional neural networks
deep learning
label-free imaging
Journal
Molecular & cellular oncology
ISSN: 2372-3556
Titre abrégé: Mol Cell Oncol
Pays: United States
ID NLM: 101642411
Informations de publication
Date de publication:
2020
2020
Historique:
received:
16
02
2020
revised:
23
02
2020
accepted:
25
02
2020
entrez:
12
5
2020
pubmed:
12
5
2020
medline:
12
5
2020
Statut:
epublish
Résumé
We recently developed and validated a bedside tissue-to-diagnosis pipeline using stimulated Raman histology (SRH), a label-free optical imaging method, and deep convolutional neural networks (CNN) in prospective clinical trial. Our CNN learned a hierarchy of interpretable histologic features found in the most common brain tumors and was able to accurately segment cancerous regions in SRH images.
Identifiants
pubmed: 32391430
doi: 10.1080/23723556.2020.1736742
pii: 1736742
pmc: PMC7199763
doi:
Types de publication
Journal Article
Langues
eng
Pagination
1736742Subventions
Organisme : NCI NIH HHS
ID : R01 CA226527
Pays : United States
Informations de copyright
© 2020 The Author(s). Published with license by Taylor & Francis Group, LLC.
Références
JAMA. 2016 Dec 13;316(22):2402-2410
pubmed: 27898976
Nature. 2017 Feb 2;542(7639):115-118
pubmed: 28117445
JAMA. 2005 Dec 28;294(24):3135-7
pubmed: 16380595
Nat Biomed Eng. 2017;1:
pubmed: 28955599
Science. 2008 Dec 19;322(5909):1857-61
pubmed: 19095943
Nat Med. 2020 Jan;26(1):52-58
pubmed: 31907460
Arch Pathol Lab Med. 2013 Dec;137(12):1723-32
pubmed: 23738764
Cancer Res. 2018 Jan 1;78(1):278-289
pubmed: 29093006
Nat Med. 2018 Sep;24(9):1337-1341
pubmed: 30104767