Your browser doesn't support javascript.
loading
Automated curation of large-scale cancer histopathology image datasets using deep learning.
Hilgers, Lars; Ghaffari Laleh, Narmin; West, Nicholas P; Westwood, Alice; Hewitt, Katherine J; Quirke, Philip; Grabsch, Heike I; Carrero, Zunamys I; Matthaei, Emylou; Loeffler, Chiara M L; Brinker, Titus J; Yuan, Tanwei; Brenner, Hermann; Brobeil, Alexander; Hoffmeister, Michael; Kather, Jakob Nikolas.
Afiliação
  • Hilgers L; Department of Medicine III, University Hospital RWTH Aachen, Aachen, Germany.
  • Ghaffari Laleh N; Else Kroener Fresenius Center for Digital Health, Medical Faculty Carl Gustav Carus, Technical University Dresden, Dresden, Germany.
  • West NP; Department of Medicine III, University Hospital RWTH Aachen, Aachen, Germany.
  • Westwood A; Else Kroener Fresenius Center for Digital Health, Medical Faculty Carl Gustav Carus, Technical University Dresden, Dresden, Germany.
  • Hewitt KJ; Pathology & Data Analytics, Leeds Institute of Medical Research at St James's, University of Leeds, Leeds, UK.
  • Quirke P; Pathology & Data Analytics, Leeds Institute of Medical Research at St James's, University of Leeds, Leeds, UK.
  • Grabsch HI; Department of Medicine III, University Hospital RWTH Aachen, Aachen, Germany.
  • Carrero ZI; Else Kroener Fresenius Center for Digital Health, Medical Faculty Carl Gustav Carus, Technical University Dresden, Dresden, Germany.
  • Matthaei E; Pathology & Data Analytics, Leeds Institute of Medical Research at St James's, University of Leeds, Leeds, UK.
  • Loeffler CML; Pathology & Data Analytics, Leeds Institute of Medical Research at St James's, University of Leeds, Leeds, UK.
  • Brinker TJ; Department of Pathology, GROW - Research Institute for Oncology and Reproduction, Maastricht University Medical Center+, Maastricht, The Netherlands.
  • Yuan T; Else Kroener Fresenius Center for Digital Health, Medical Faculty Carl Gustav Carus, Technical University Dresden, Dresden, Germany.
  • Brenner H; Else Kroener Fresenius Center for Digital Health, Medical Faculty Carl Gustav Carus, Technical University Dresden, Dresden, Germany.
  • Brobeil A; Else Kroener Fresenius Center for Digital Health, Medical Faculty Carl Gustav Carus, Technical University Dresden, Dresden, Germany.
  • Hoffmeister M; Digital Biomarkers for Oncology Group, German Cancer Research Center (DKFZ), Heidelberg, Germany.
  • Kather JN; Division of Clinical Epidemiology and Aging Research, German Cancer Research Center (DKFZ), Heidelberg, Germany.
Histopathology ; 84(7): 1139-1153, 2024 Jun.
Article em En | MEDLINE | ID: mdl-38409878
ABSTRACT

BACKGROUND:

Artificial intelligence (AI) has numerous applications in pathology, supporting diagnosis and prognostication in cancer. However, most AI models are trained on highly selected data, typically one tissue slide per patient. In reality, especially for large surgical resection specimens, dozens of slides can be available for each patient. Manually sorting and labelling whole-slide images (WSIs) is a very time-consuming process, hindering the direct application of AI on the collected tissue samples from large cohorts. In this study we addressed this issue by developing a deep-learning (DL)-based method for automatic curation of large pathology datasets with several slides per patient.

METHODS:

We collected multiple large multicentric datasets of colorectal cancer histopathological slides from the United Kingdom (FOXTROT, N = 21,384 slides; CR07, N = 7985 slides) and Germany (DACHS, N = 3606 slides). These datasets contained multiple types of tissue slides, including bowel resection specimens, endoscopic biopsies, lymph node resections, immunohistochemistry-stained slides, and tissue microarrays. We developed, trained, and tested a deep convolutional neural network model to predict the type of slide from the slide overview (thumbnail) image. The primary statistical endpoint was the macro-averaged area under the receiver operating curve (AUROCs) for detection of the type of slide.

RESULTS:

In the primary dataset (FOXTROT), with an AUROC of 0.995 [95% confidence interval [CI] 0.994-0.996] the algorithm achieved a high classification performance and was able to accurately predict the type of slide from the thumbnail image alone. In the two external test cohorts (CR07, DACHS) AUROCs of 0.982 [95% CI 0.979-0.985] and 0.875 [95% CI 0.864-0.887] were observed, which indicates the generalizability of the trained model on unseen datasets. With a confidence threshold of 0.95, the model reached an accuracy of 94.6% (7331 classified cases) in CR07 and 85.1% (2752 classified cases) for the DACHS cohort.

CONCLUSION:

Our findings show that using the low-resolution thumbnail image is sufficient to accurately classify the type of slide in digital pathology. This can support researchers to make the vast resource of existing pathology archives accessible to modern AI models with only minimal manual annotations.
Assuntos
Palavras-chave

Texto completo: 1 Base de dados: MEDLINE Assunto principal: Neoplasias Colorretais / Aprendizado Profundo Limite: Humans Idioma: En Ano de publicação: 2024 Tipo de documento: Article

Texto completo: 1 Base de dados: MEDLINE Assunto principal: Neoplasias Colorretais / Aprendizado Profundo Limite: Humans Idioma: En Ano de publicação: 2024 Tipo de documento: Article