Tens of images can suffice to train neural networks for malignant leukocyte detection

Abstract Convolutional neural networks (CNNs) excel as powerful tools for biomedical image classification. It is commonly assumed that training CNNs requires large amounts of annotated data. This is a bottleneck in many medical applications where annotation relies on expert knowledge. Here, we analy...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Jens P. E. Schouten, Christian Matek, Luuk F. P. Jacobs, Michèle C. Buck, Dragan Bošnački, Carsten Marr
Formato: article
Lenguaje:EN
Publicado: Nature Portfolio 2021
Materias:
R
Q
Acceso en línea:https://doaj.org/article/61eba0d963d64a72884cde1b14ed2b2a
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
id oai:doaj.org-article:61eba0d963d64a72884cde1b14ed2b2a
record_format dspace
spelling oai:doaj.org-article:61eba0d963d64a72884cde1b14ed2b2a2021-12-02T15:51:14ZTens of images can suffice to train neural networks for malignant leukocyte detection10.1038/s41598-021-86995-52045-2322https://doaj.org/article/61eba0d963d64a72884cde1b14ed2b2a2021-04-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-86995-5https://doaj.org/toc/2045-2322Abstract Convolutional neural networks (CNNs) excel as powerful tools for biomedical image classification. It is commonly assumed that training CNNs requires large amounts of annotated data. This is a bottleneck in many medical applications where annotation relies on expert knowledge. Here, we analyze the binary classification performance of a CNN on two independent cytomorphology datasets as a function of training set size. Specifically, we train a sequential model to discriminate non-malignant leukocytes from blast cells, whose appearance in the peripheral blood is a hallmark of leukemia. We systematically vary training set size, finding that tens of training images suffice for a binary classification with an ROC-AUC over 90%. Saliency maps and layer-wise relevance propagation visualizations suggest that the network learns to increasingly focus on nuclear structures of leukocytes as the number of training images is increased. A low dimensional tSNE representation reveals that while the two classes are separated already for a few training images, the distinction between the classes becomes clearer when more training images are used. To evaluate the performance in a multi-class problem, we annotated single-cell images from a acute lymphoblastic leukemia dataset into six different hematopoietic classes. Multi-class prediction suggests that also here few single-cell images suffice if differences between morphological classes are large enough. The incorporation of deep learning algorithms into clinical practice has the potential to reduce variability and cost, democratize usage of expertise, and allow for early detection of disease onset and relapse. Our approach evaluates the performance of a deep learning based cytology classifier with respect to size and complexity of the training data and the classification task.Jens P. E. SchoutenChristian MatekLuuk F. P. JacobsMichèle C. BuckDragan BošnačkiCarsten MarrNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-8 (2021)
institution DOAJ
collection DOAJ
language EN
topic Medicine
R
Science
Q
spellingShingle Medicine
R
Science
Q
Jens P. E. Schouten
Christian Matek
Luuk F. P. Jacobs
Michèle C. Buck
Dragan Bošnački
Carsten Marr
Tens of images can suffice to train neural networks for malignant leukocyte detection
description Abstract Convolutional neural networks (CNNs) excel as powerful tools for biomedical image classification. It is commonly assumed that training CNNs requires large amounts of annotated data. This is a bottleneck in many medical applications where annotation relies on expert knowledge. Here, we analyze the binary classification performance of a CNN on two independent cytomorphology datasets as a function of training set size. Specifically, we train a sequential model to discriminate non-malignant leukocytes from blast cells, whose appearance in the peripheral blood is a hallmark of leukemia. We systematically vary training set size, finding that tens of training images suffice for a binary classification with an ROC-AUC over 90%. Saliency maps and layer-wise relevance propagation visualizations suggest that the network learns to increasingly focus on nuclear structures of leukocytes as the number of training images is increased. A low dimensional tSNE representation reveals that while the two classes are separated already for a few training images, the distinction between the classes becomes clearer when more training images are used. To evaluate the performance in a multi-class problem, we annotated single-cell images from a acute lymphoblastic leukemia dataset into six different hematopoietic classes. Multi-class prediction suggests that also here few single-cell images suffice if differences between morphological classes are large enough. The incorporation of deep learning algorithms into clinical practice has the potential to reduce variability and cost, democratize usage of expertise, and allow for early detection of disease onset and relapse. Our approach evaluates the performance of a deep learning based cytology classifier with respect to size and complexity of the training data and the classification task.
format article
author Jens P. E. Schouten
Christian Matek
Luuk F. P. Jacobs
Michèle C. Buck
Dragan Bošnački
Carsten Marr
author_facet Jens P. E. Schouten
Christian Matek
Luuk F. P. Jacobs
Michèle C. Buck
Dragan Bošnački
Carsten Marr
author_sort Jens P. E. Schouten
title Tens of images can suffice to train neural networks for malignant leukocyte detection
title_short Tens of images can suffice to train neural networks for malignant leukocyte detection
title_full Tens of images can suffice to train neural networks for malignant leukocyte detection
title_fullStr Tens of images can suffice to train neural networks for malignant leukocyte detection
title_full_unstemmed Tens of images can suffice to train neural networks for malignant leukocyte detection
title_sort tens of images can suffice to train neural networks for malignant leukocyte detection
publisher Nature Portfolio
publishDate 2021
url https://doaj.org/article/61eba0d963d64a72884cde1b14ed2b2a
work_keys_str_mv AT jenspeschouten tensofimagescansufficetotrainneuralnetworksformalignantleukocytedetection
AT christianmatek tensofimagescansufficetotrainneuralnetworksformalignantleukocytedetection
AT luukfpjacobs tensofimagescansufficetotrainneuralnetworksformalignantleukocytedetection
AT michelecbuck tensofimagescansufficetotrainneuralnetworksformalignantleukocytedetection
AT draganbosnacki tensofimagescansufficetotrainneuralnetworksformalignantleukocytedetection
AT carstenmarr tensofimagescansufficetotrainneuralnetworksformalignantleukocytedetection
_version_ 1718385635030466560