HistoMIL: A Python package for training multiple instance learning models on histopathology slides.
Artificial intelligence
Histology
Machine learning
Journal
iScience
ISSN: 2589-0042
Titre abrégé: iScience
Pays: United States
ID NLM: 101724038
Informations de publication
Date de publication:
20 Oct 2023
20 Oct 2023
Historique:
received:
01
06
2023
revised:
21
08
2023
accepted:
25
09
2023
medline:
20
10
2023
pubmed:
20
10
2023
entrez:
20
10
2023
Statut:
epublish
Résumé
Hematoxylin and eosin (H&E) stained slides are widely used in disease diagnosis. Remarkable advances in deep learning have made it possible to detect complex molecular patterns in these histopathology slides, suggesting automated approaches could help inform pathologists' decisions. Multiple instance learning (MIL) algorithms have shown promise in this context, outperforming transfer learning (TL) methods for various tasks, but their implementation and usage remains complex. We introduce HistoMIL, a Python package designed to streamline the implementation, training and inference process of MIL-based algorithms for computational pathologists and biomedical researchers. It integrates a self-supervised learning module for feature encoding, and a full pipeline encompassing TL and three MIL algorithms: ABMIL, DSMIL, and TransMIL. The PyTorch Lightning framework enables effortless customization and algorithm implementation. We illustrate HistoMIL's capabilities by building predictive models for 2,487 cancer hallmark genes on breast cancer histology slides, achieving AUROC performances of up to 85%.
Identifiants
pubmed: 37860768
doi: 10.1016/j.isci.2023.108073
pii: S2589-0042(23)02150-8
pmc: PMC10583115
doi:
Types de publication
Journal Article
Langues
eng
Pagination
108073Informations de copyright
© 2023 The Author(s).
Déclaration de conflit d'intérêts
The authors declare no competing interests.
Références
NPJ Breast Cancer. 2023 Apr 14;9(1):25
pubmed: 37059742
Front Biosci. 1998 Apr 27;3:d447-8
pubmed: 9556498
Comput Intell Neurosci. 2022 Mar 19;2022:1820777
pubmed: 35345799
Cancer Res. 2014 Jul 1;74(13):3630-42
pubmed: 24786789
Cancers (Basel). 2019 Oct 28;11(11):
pubmed: 31661863
Commun Med (Lond). 2022 Sep 24;2:120
pubmed: 36168445
J Biol Chem. 2018 Jan 12;293(2):484-496
pubmed: 29162720
Cell Syst. 2015 Dec 23;1(6):417-425
pubmed: 26771021
Anal Chem. 2023 Feb 14;95(6):3291-3299
pubmed: 36724070
J Clin Oncol. 2000 Mar;18(5):1135-49
pubmed: 10694567
Curr Biol. 2011 Aug 23;21(16):1356-65
pubmed: 21820309
Annu Int Conf IEEE Eng Med Biol Soc. 2022 Jul;2022:3068-3071
pubmed: 36085965
Breast Cancer Res Treat. 2012 Jun;133(3):909-16
pubmed: 22076481
Nat Med. 2018 Oct;24(10):1559-1567
pubmed: 30224757
Nat Biomed Eng. 2021 Jun;5(6):555-570
pubmed: 33649564
N Engl J Med. 2016 Sep 22;375(12):1109-12
pubmed: 27653561
Sci Rep. 2021 Jun 2;11(1):11579
pubmed: 34078928
Cancer Res. 2002 Sep 15;62(18):5358-64
pubmed: 12235007
Nat Med. 2019 Jul;25(7):1054-1056
pubmed: 31160815
Nat Commun. 2020 Aug 3;11(1):3877
pubmed: 32747659
Nature. 2015 May 28;521(7553):436-44
pubmed: 26017442
Med Biol Eng Comput. 2019 Sep;57(9):2027-2043
pubmed: 31346949
PLoS Comput Biol. 2020 Oct 19;16(10):e1008349
pubmed: 33075075
J Pathol Inform. 2013 Sep 27;4:27
pubmed: 24244884
Conf Comput Vis Pattern Recognit Workshops. 2021 Jun;2021:14318-14328
pubmed: 35047230
Med Image Anal. 2019 Dec;58:101563
pubmed: 31561183
Nucleic Acids Res. 2010 Jul;38(Web Server issue):W214-20
pubmed: 20576703
Nucleic Acids Res. 2016 May 5;44(8):e71
pubmed: 26704973
Diagnostics (Basel). 2021 Aug 03;11(8):
pubmed: 34441338
Nat Genet. 2013 Oct;45(10):1113-20
pubmed: 24071849
Med Image Anal. 2020 Oct;65:101789
pubmed: 32739769
J Digit Imaging. 2022 Dec;35(6):1719-1737
pubmed: 35995898
Mol Biol Cell. 2015 Oct 1;26(19):3424-38
pubmed: 26246606
Nat Commun. 2021 Feb 19;12(1):1193
pubmed: 33608558