Development and evaluation of deep learning-based segmentation of histologic structures in the kidney cortex with multiple histologic stains.
computerized morphologic assessment
deep learning
digital pathology
kidney histologic primitives
large-scale tissue interrogation
renal biopsy interpretation
Journal
Kidney international
ISSN: 1523-1755
Titre abrégé: Kidney Int
Pays: United States
ID NLM: 0323470
Informations de publication
Date de publication:
01 2021
01 2021
Historique:
received:
27
12
2019
revised:
29
06
2020
accepted:
24
07
2020
pubmed:
25
8
2020
medline:
22
6
2021
entrez:
25
8
2020
Statut:
ppublish
Résumé
The application of deep learning for automated segmentation (delineation of boundaries) of histologic primitives (structures) from whole slide images can facilitate the establishment of novel protocols for kidney biopsy assessment. Here, we developed and validated deep learning networks for the segmentation of histologic structures on kidney biopsies and nephrectomies. For development, we examined 125 biopsies for Minimal Change Disease collected across 29 NEPTUNE enrolling centers along with 459 whole slide images stained with Hematoxylin & Eosin (125), Periodic Acid Schiff (125), Silver (102), and Trichrome (107) divided into training, validation and testing sets (ratio 6:1:3). Histologic structures were manually segmented (30048 total annotations) by five nephropathologists. Twenty deep learning models were trained with optimal digital magnification across the structures and stains. Periodic Acid Schiff-stained whole slide images yielded the best concordance between pathologists and deep learning segmentation across all structures (F-scores: 0.93 for glomerular tufts, 0.94 for glomerular tuft plus Bowman's capsule, 0.91 for proximal tubules, 0.93 for distal tubular segments, 0.81 for peritubular capillaries, and 0.85 for arteries and afferent arterioles). Optimal digital magnifications were 5X for glomerular tuft/tuft plus Bowman's capsule, 10X for proximal/distal tubule, arteries and afferent arterioles, and 40X for peritubular capillaries. Silver stained whole slide images yielded the worst deep learning performance. Thus, this largest study to date adapted deep learning for the segmentation of kidney histologic structures across multiple stains and pathology laboratories. All data used for training and testing and a detailed online tutorial will be publicly available.
Identifiants
pubmed: 32835732
pii: S0085-2538(20)30962-5
doi: 10.1016/j.kint.2020.07.044
pmc: PMC8414393
mid: NIHMS1715905
pii:
doi:
Substances chimiques
Coloring Agents
0
Types de publication
Journal Article
Research Support, N.I.H., Extramural
Research Support, Non-U.S. Gov't
Research Support, U.S. Gov't, Non-P.H.S.
Langues
eng
Sous-ensembles de citation
IM
Pagination
86-101Subventions
Organisme : NCI NIH HHS
ID : R01 CA216579
Pays : United States
Organisme : NCRR NIH HHS
ID : C06 RR012463
Pays : United States
Organisme : NCI NIH HHS
ID : U24 CA199374
Pays : United States
Organisme : NIBIB NIH HHS
ID : R43 EB028736
Pays : United States
Organisme : NCI NIH HHS
ID : U01 CA239055
Pays : United States
Organisme : NIDDK NIH HHS
ID : T32 DK007470
Pays : United States
Organisme : NCI NIH HHS
ID : R01 CA220581
Pays : United States
Organisme : NCI NIH HHS
ID : R01 CA202752
Pays : United States
Organisme : NCI NIH HHS
ID : R01 CA208236
Pays : United States
Organisme : NCI NIH HHS
ID : U01 CA248226
Pays : United States
Organisme : BLRD VA
ID : I01 BX004121
Pays : United States
Investigateurs
J Sedor
(J)
K Dell
(K)
M Schachere
(M)
J Negrey
(J)
K Lemley
(K)
E Lim
(E)
T Srivastava
(T)
A Garrett
(A)
C Sethna
(C)
K Laurent
(K)
G Appel
(G)
M Toledo
(M)
L Barisoni
(L)
L Greenbaum
(L)
C Wang
(C)
C Kang
(C)
S Adler
(S)
C Nast
(C)
J LaPage
(J)
John H Stroger
(JH)
A Athavale
(A)
M Itteera
(M)
A Neu
(A)
S Boynton
(S)
F Fervenza
(F)
M Hogan
(M)
J Lieske
(J)
V Chernitskiy
(V)
F Kaskel
(F)
N Kumar
(N)
P Flynn
(P)
J Kopp
(J)
J Blake
(J)
H Trachtman
(H)
O Zhdanova
(O)
F Modersitzki
(F)
S Vento
(S)
R Lafayette
(R)
K Mehta
(K)
C Gadegbeku
(C)
D Johnstone
(D)
S Quinn-Boyle
(S)
D Cattran
(D)
M Hladunewich
(M)
H Reich
(H)
P Ling
(P)
M Romano
(M)
A Fornoni
(A)
C Bidot
(C)
M Kretzler
(M)
D Gipson
(D)
A Williams
(A)
J LaVigne
(J)
V Derebail
(V)
K Gibson
(K)
A Froment
(A)
S Grubbs
(S)
L Holzman
(L)
K Meyers
(K)
K Kallem
(K)
J Lalli
(J)
K Sambandam
(K)
Z Wang
(Z)
M Rogers
(M)
A Jefferson
(A)
S Hingorani
(S)
K Tuttle
(K)
M Bray
(M)
M Kelton
(M)
A Cooper
(A)
B Freedman
(B)
J J Lin
(JJ)
Commentaires et corrections
Type : CommentIn
Informations de copyright
Copyright © 2020 International Society of Nephrology. All rights reserved.
Références
Kidney Int. 2018 May;93(5):1175-1182
pubmed: 29273332
Acad Radiol. 2004 Feb;11(2):178-89
pubmed: 14974593
Clin Kidney J. 2017 Apr;10(2):176-187
pubmed: 28584625
IEEE Trans Med Imaging. 2019 Oct;38(10):2293-2302
pubmed: 30762541
Med Image Anal. 2016 Oct;33:170-175
pubmed: 27423409
Clin J Am Soc Nephrol. 2016 Feb 5;11(2):354-62
pubmed: 26339068
J Am Soc Nephrol. 2018 Aug;29(8):2213-2224
pubmed: 29866798
PLoS One. 2016 Jul 06;11(7):e0158765
pubmed: 27383068
BMC Cancer. 2018 May 30;18(1):610
pubmed: 29848291
J Am Soc Nephrol. 2019 Oct;30(10):1953-1967
pubmed: 31488606
Can J Urol. 2016 Feb;23(1):8121-6
pubmed: 26892051
Pol J Pathol. 2003;54(3):163-9
pubmed: 14703282
Kidney Int Rep. 2018 Jan 11;3(2):464-475
pubmed: 29725651
J Am Soc Nephrol. 2019 Oct;30(10):1968-1979
pubmed: 31488607
Sci Rep. 2019 Jul 19;9(1):10509
pubmed: 31324828
Contrib Nephrol. 2011;169:37-50
pubmed: 21252510
Am J Hypertens. 2019 Jan 1;32(1):45-53
pubmed: 30358804
Sci Rep. 2019 Mar 4;9(1):3358
pubmed: 30833650
Nat Med. 2019 Aug;25(8):1301-1309
pubmed: 31308507
Nephrology (Carlton). 2015 Jan;20(1):11-7
pubmed: 25230383
JAMA. 2017 Dec 12;318(22):2199-2210
pubmed: 29234806
Nephrol Dial Transplant. 2018 Feb 1;33(2):310-318
pubmed: 28339906
Kidney Int. 2020 Jul;98(1):65-75
pubmed: 32475607
Nat Rev Clin Oncol. 2019 Nov;16(11):703-715
pubmed: 31399699
Front Public Health. 2017 Nov 20;5:307
pubmed: 29209603
Curr Opin Nephrol Hypertens. 2020 May;29(3):265-272
pubmed: 32205581
Kidney Int. 2018 May;93(5):1043-1044
pubmed: 29680021
PLoS One. 2016 Jun 10;11(6):e0157148
pubmed: 27285824
Pediatr Nephrol. 2016 May;31(5):693-706
pubmed: 26208584
Nat Mach Intell. 2019 Feb;1(2):112-119
pubmed: 31187088
Am J Surg Pathol. 2014 Jan;38(1):128-37
pubmed: 24145650
Mod Pathol. 2016 Jul;29(7):671-84
pubmed: 27102348
J Am Soc Nephrol. 2014 May;25(5):1118-29
pubmed: 24357669
Med Image Anal. 2017 Dec;42:60-88
pubmed: 28778026
J Pathol Inform. 2016 Jul 26;7:29
pubmed: 27563488
Kidney Int Rep. 2019 Apr 15;4(7):955-962
pubmed: 31317118
IEEE Trans Med Imaging. 2018 Dec;37(12):2718-2728
pubmed: 29994669
Clin J Am Soc Nephrol. 2013 Aug;8(8):1449-59
pubmed: 23393107
Am J Kidney Dis. 1992 Jul;20(1):1-17
pubmed: 1621674
IEEE Trans Med Imaging. 2016 May;35(5):1285-98
pubmed: 26886976
PLoS Med. 2019 Jan 24;16(1):e1002730
pubmed: 30677016
Nephrol Dial Transplant. 2001 Jun;16(6):1163-9
pubmed: 11390715
Lupus. 2017 Oct;26(11):1205-1211
pubmed: 28478696
Nat Rev Nephrol. 2020 Jan;16(1):4-6
pubmed: 31597956
JCO Clin Cancer Inform. 2019 Apr;3:1-7
pubmed: 30990737
Curr Opin Nephrol Hypertens. 2017 Nov;26(6):450-459
pubmed: 28858910
Ann Intern Med. 1993 Oct 15;119(8):805-11
pubmed: 8379602