A deep-learning model for transforming the style of tissue images from cryosectioned to formalin-fixed and paraffin-embedded

Kutsev Bengisu Ozyoruk; Sermet Can; Berkan Darbaz; Kayhan Başak; Derya Demir; Guliz Irem Gokceler; Gurdeniz Serin; Uguray Payam Hacisalihoglu; Emirhan Kurtuluş; Ming Y Lu; Tiffany Y Chen; Drew F K Williamson; Funda Yılmaz; Faisal Mahmood; Mehmet Turan

doi:10.1038/s41551-022-00952-9

A deep-learning model for transforming the style of tissue images from cryosectioned to formalin-fixed and paraffin-embedded

Nat Biomed Eng. 2022 Dec;6(12):1407-1419. doi: 10.1038/s41551-022-00952-9. Epub 2022 Dec 23.

Authors

Kutsev Bengisu Ozyoruk^{1

2

3}, Sermet Can^{1

3}, Berkan Darbaz^{1

3

4}, Kayhan Başak⁵, Derya Demir⁶, Guliz Irem Gokceler^{1

3}, Gurdeniz Serin⁶, Uguray Payam Hacisalihoglu⁷, Emirhan Kurtuluş⁸, Ming Y Lu⁹, Tiffany Y Chen², Drew F K Williamson², Funda Yılmaz⁶, Faisal Mahmood^#^{10

11

12}, Mehmet Turan^#^{13

14}

Affiliations

¹ Department of Computer Engineering, Bogazici University, Istanbul, Turkey.
² Department of Pathology, Brigham and Women's Hospital and Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA.
³ Institute of Biomedical Engineering, Bogazici University, Istanbul, Turkey.
⁴ Virasoft Corporation, New York, NY, USA.
⁵ Department of Pathology, Sağlık Bilimleri University, Kartal Dr.Lütfi Kırdar City Hospital, Istanbul, Turkey.
⁶ Faculty of Medicine, Department of Pathology, Ege University, Izmir, Turkey.
⁷ Pathology Department, Istanbul Yeni Yuzyil University Medical Faculty, Gaziosmanpasa Hospital, Izmir, Turkey.
⁸ Stanford University, Stanford, CA, USA.
⁹ Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, MA, USA.
¹⁰ Department of Pathology, Brigham and Women's Hospital and Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA. faisalmahmood@bwh.harvard.edu.
¹¹ Cancer Data Science Program, Dana-Farber Cancer Institute, Boston, MA, USA. faisalmahmood@bwh.harvard.edu.
¹² Harvard Data Science Initiative, Harvard University, Cambridge, MA, USA. faisalmahmood@bwh.harvard.edu.
¹³ Department of Computer Engineering, Bogazici University, Istanbul, Turkey. mehmet.turan@boun.edu.tr.
¹⁴ Institute of Biomedical Engineering, Bogazici University, Istanbul, Turkey. mehmet.turan@boun.edu.tr.

^# Contributed equally.

PMID: 36564629
DOI: 10.1038/s41551-022-00952-9

Abstract

Histological artefacts in cryosectioned tissue can hinder rapid diagnostic assessments during surgery. Formalin-fixed and paraffin-embedded (FFPE) tissue provides higher quality slides, but the process for obtaining them is laborious (typically lasting 12-48 h) and hence unsuitable for intra-operative use. Here we report the development and performance of a deep-learning model that improves the quality of cryosectioned whole-slide images by transforming them into the style of whole-slide FFPE tissue within minutes. The model consists of a generative adversarial network incorporating an attention mechanism that rectifies cryosection artefacts and a self-regularization constraint between the cryosectioned and FFPE images for the preservation of clinically relevant features. Transformed FFPE-style images of gliomas and of non-small-cell lung cancers from a dataset independent from that used to train the model improved the rates of accurate tumour subtyping by pathologists.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Carcinoma, Non-Small-Cell Lung*
Deep Learning*
Formaldehyde
Humans
Lung Neoplasms*
Paraffin Embedding / methods

Substances

Formaldehyde