Medulloblastoma tumor classification using deep transfer learning with multi-scale EfficientNets

TY - JOUR

T1 - Medulloblastoma tumor classification using deep transfer learning with multi-scale EfficientNets

AU - Bengs, Marcel

AU - Bockmayr, Michael

AU - Schüller, Ulrich

AU - Schlaefer, Alexander

N1 - Funding Information: Acknowledgments. This work was partially supported by the Hamburg University of Technology i3 initiative. Publisher Copyright: © COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only. Copyright: Copyright 2021 Elsevier B.V., All rights reserved.

PY - 2021

Y1 - 2021

N2 - Medulloblastoma (MB) is the most common malignant brain tumor in childhood. The diagnosis is generally based on the microscopic evaluation of histopathological tissue slides. However, visual-only assessment of histopathological patterns is a tedious and time-consuming task and is also affected by observer variability. Hence, automated MB tumor classification could assist pathologists by promoting consistency and robust quantification. Recently, convolutional neural networks (CNNs) have been proposed for this task, while transfer learning has shown promising results. In this work, we propose an end-to-end MB tumor classification and explore transfer learning with various input sizes and matching network dimensions. We focus on differentiating between the histological subtypes classic and desmoplastic/nodular. For this purpose, we systematically evaluate recently proposed EfficientNets, which uniformly scale all dimensions of a CNN. Using a data set with 161 cases, we demonstrate that pre-trained EfficientNets with larger input resolutions lead to significant performance improvements compared to commonly used pre-trained CNN architectures. Also, we highlight the importance of transfer learning, when using such large architectures. Overall, our best performing method achieves an F1-Score of 80.1%.

AB - Medulloblastoma (MB) is the most common malignant brain tumor in childhood. The diagnosis is generally based on the microscopic evaluation of histopathological tissue slides. However, visual-only assessment of histopathological patterns is a tedious and time-consuming task and is also affected by observer variability. Hence, automated MB tumor classification could assist pathologists by promoting consistency and robust quantification. Recently, convolutional neural networks (CNNs) have been proposed for this task, while transfer learning has shown promising results. In this work, we propose an end-to-end MB tumor classification and explore transfer learning with various input sizes and matching network dimensions. We focus on differentiating between the histological subtypes classic and desmoplastic/nodular. For this purpose, we systematically evaluate recently proposed EfficientNets, which uniformly scale all dimensions of a CNN. Using a data set with 161 cases, we demonstrate that pre-trained EfficientNets with larger input resolutions lead to significant performance improvements compared to commonly used pre-trained CNN architectures. Also, we highlight the importance of transfer learning, when using such large architectures. Overall, our best performing method achieves an F1-Score of 80.1%.

UR - http://www.scopus.com/inward/record.url?scp=85103300797&partnerID=8YFLogxK

U2 - 10.1117/12.2580717

DO - 10.1117/12.2580717

M3 - Conference article in journal

AN - SCOPUS:85103300797

JO - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

JF - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

SN - 1605-7422

T2 - Medical Imaging 2021: Digital Pathology

Y2 - 15 February 2021 through 19 February 2021

ER -