Papillary thyroid carcinoma whole-slide images as a basis for deep learning

Objectives. Morphological analysis of papillary thyroid cancer is a cornerstone for further treatment planning. Traditional and neural network methods of extracting parts of images are used to automate the analysis. It is necessary to prepare a set of data for teaching neural networks to develop a s...

Full description

Saved in:
Bibliographic Details
Main Authors: M. V. Fridman, A. A. Kosareva, E. V. Snezhko, P. V. Kamlach, V. A. Kovalev
Format: Article
Language:Russian
Published: National Academy of Sciences of Belarus, the United Institute of Informatics Problems 2023-06-01
Series:Informatika
Subjects:
Online Access:https://inf.grid.by/jour/article/view/1241
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849771461232820224
author M. V. Fridman
A. A. Kosareva
E. V. Snezhko
P. V. Kamlach
V. A. Kovalev
author_facet M. V. Fridman
A. A. Kosareva
E. V. Snezhko
P. V. Kamlach
V. A. Kovalev
author_sort M. V. Fridman
collection DOAJ
description Objectives. Morphological analysis of papillary thyroid cancer is a cornerstone for further treatment planning. Traditional and neural network methods of extracting parts of images are used to automate the analysis. It is necessary to prepare a set of data for teaching neural networks to develop a system of similar anatomical region in the histopathological image. Authors discuss the second selection of signs for the marking of histological images, methodological approaches to dissect whole-slide images, how to prepare raw data for a future analysis. The influence of the representative size of the fragment of the full-to-suction image of papillary thyroid cancer on the accuracy of the classification of trained neural network EfficientNetB0 is conducted. The analysis of the resulting results is carried out, the weaknesses of the use of fragments of images of different representative size and the cause of the unsatisfactory accuracy of the classification on large increase are evaluated.Materials and methods. Histopathological whole-slide imaged of 129 patients were used. Histological micropreparations containing elements of a tumor and surrounding tissue were scanned in the Aperio AT2 (Leica Biosystems, Germany) apparatus with maximum resolution. The marking was carried out in the ASAP software package. To choose the optimal representative size of the fragment the problem of classification was solved using the pre-study neural network EfficientNetB0.Results. A methodology for preparing a database of histopathological images of papillary thyroid cancer was proposed. Experiments were conducted to determine the optimal representative size of the image fragment. The best result of the accuracy of determining the class of test sample showed the size of a representative fragment as 394.32×394.32 microns.Conclusion. The analysis of the influence of the representative sizes of fragments of histopathological images showed the problems in solving the classification tasks because of cutting and staining images specifics, morphological complex and textured differences in the images of the same class. At the same time, it was determined that the task of preparing a set of data for training neural network to solve the problem of finding invasion of vessels in a histopathological image is not trivial and it requires additional stages of data preparation.
format Article
id doaj-art-bfac95fe845d43d9a7f0db9b6029a3d2
institution DOAJ
issn 1816-0301
language Russian
publishDate 2023-06-01
publisher National Academy of Sciences of Belarus, the United Institute of Informatics Problems
record_format Article
series Informatika
spelling doaj-art-bfac95fe845d43d9a7f0db9b6029a3d22025-08-20T03:02:37ZrusNational Academy of Sciences of Belarus, the United Institute of Informatics ProblemsInformatika1816-03012023-06-01202283810.37661/1816-0301-2023-20-2-28-381034Papillary thyroid carcinoma whole-slide images as a basis for deep learningM. V. Fridman0A. A. Kosareva1E. V. Snezhko2P. V. Kamlach3V. A. Kovalev4Minsk City Clinical Oncology Center; Republican Centre for Thyroid TumoursBelarusian State University of Informatics and RadioelectronicsThe United Institute of Informatics Problems of the National Academy of Sciences of BelarusBelarusian State University of Informatics and RadioelectronicsThe United Institute of Informatics Problems of the National Academy of Sciences of BelarusObjectives. Morphological analysis of papillary thyroid cancer is a cornerstone for further treatment planning. Traditional and neural network methods of extracting parts of images are used to automate the analysis. It is necessary to prepare a set of data for teaching neural networks to develop a system of similar anatomical region in the histopathological image. Authors discuss the second selection of signs for the marking of histological images, methodological approaches to dissect whole-slide images, how to prepare raw data for a future analysis. The influence of the representative size of the fragment of the full-to-suction image of papillary thyroid cancer on the accuracy of the classification of trained neural network EfficientNetB0 is conducted. The analysis of the resulting results is carried out, the weaknesses of the use of fragments of images of different representative size and the cause of the unsatisfactory accuracy of the classification on large increase are evaluated.Materials and methods. Histopathological whole-slide imaged of 129 patients were used. Histological micropreparations containing elements of a tumor and surrounding tissue were scanned in the Aperio AT2 (Leica Biosystems, Germany) apparatus with maximum resolution. The marking was carried out in the ASAP software package. To choose the optimal representative size of the fragment the problem of classification was solved using the pre-study neural network EfficientNetB0.Results. A methodology for preparing a database of histopathological images of papillary thyroid cancer was proposed. Experiments were conducted to determine the optimal representative size of the image fragment. The best result of the accuracy of determining the class of test sample showed the size of a representative fragment as 394.32×394.32 microns.Conclusion. The analysis of the influence of the representative sizes of fragments of histopathological images showed the problems in solving the classification tasks because of cutting and staining images specifics, morphological complex and textured differences in the images of the same class. At the same time, it was determined that the task of preparing a set of data for training neural network to solve the problem of finding invasion of vessels in a histopathological image is not trivial and it requires additional stages of data preparation.https://inf.grid.by/jour/article/view/1241medical imagingconvolutional neural networkdeep learningcomputer-aided diagnosispapillary thyroid cancercancer architectonics
spellingShingle M. V. Fridman
A. A. Kosareva
E. V. Snezhko
P. V. Kamlach
V. A. Kovalev
Papillary thyroid carcinoma whole-slide images as a basis for deep learning
Informatika
medical imaging
convolutional neural network
deep learning
computer-aided diagnosis
papillary thyroid cancer
cancer architectonics
title Papillary thyroid carcinoma whole-slide images as a basis for deep learning
title_full Papillary thyroid carcinoma whole-slide images as a basis for deep learning
title_fullStr Papillary thyroid carcinoma whole-slide images as a basis for deep learning
title_full_unstemmed Papillary thyroid carcinoma whole-slide images as a basis for deep learning
title_short Papillary thyroid carcinoma whole-slide images as a basis for deep learning
title_sort papillary thyroid carcinoma whole slide images as a basis for deep learning
topic medical imaging
convolutional neural network
deep learning
computer-aided diagnosis
papillary thyroid cancer
cancer architectonics
url https://inf.grid.by/jour/article/view/1241
work_keys_str_mv AT mvfridman papillarythyroidcarcinomawholeslideimagesasabasisfordeeplearning
AT aakosareva papillarythyroidcarcinomawholeslideimagesasabasisfordeeplearning
AT evsnezhko papillarythyroidcarcinomawholeslideimagesasabasisfordeeplearning
AT pvkamlach papillarythyroidcarcinomawholeslideimagesasabasisfordeeplearning
AT vakovalev papillarythyroidcarcinomawholeslideimagesasabasisfordeeplearning