Cargando…
Automated Estimation of Mammary Gland Content Ratio Using Regression Deep Convolutional Neural Network and the Effectiveness in Clinical Practice as Explainable Artificial Intelligence
SIMPLE SUMMARY: In Japan, a guideline for breast types, called “breast composition,” was recently developed based on BI-RADS. The Japanese guidelines are characterized using a continuous value called the mammary gland content ratio, calculated using the density of the pectoralis muscle as an indicat...
Autores principales: | , , , , , , , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
MDPI
2023
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10216583/ https://www.ncbi.nlm.nih.gov/pubmed/37345132 http://dx.doi.org/10.3390/cancers15102794 |
_version_ | 1785048333549568000 |
---|---|
author | Kai, Chiharu Ishizuka, Sachi Otsuka, Tsunehiro Nara, Miyako Kondo, Satoshi Futamura, Hitoshi Kodama, Naoki Kasai, Satoshi |
author_facet | Kai, Chiharu Ishizuka, Sachi Otsuka, Tsunehiro Nara, Miyako Kondo, Satoshi Futamura, Hitoshi Kodama, Naoki Kasai, Satoshi |
author_sort | Kai, Chiharu |
collection | PubMed |
description | SIMPLE SUMMARY: In Japan, a guideline for breast types, called “breast composition,” was recently developed based on BI-RADS. The Japanese guidelines are characterized using a continuous value called the mammary gland content ratio, calculated using the density of the pectoralis muscle as an indicator to determine breast composition. Discriminative DCNN has been developed conventionally to classify breast composition; however, it could encounter two-step errors or more (e.g., estimating “Fatty” as “Heterogeneous dense”). We developed a regression DCNN based on the mammary gland content ratio defined in the Japanese guideline to solve the above problem, followed by automated breast composition classification based on the continuous value. We also examined the usefulness of the continuous value of the mammary gland content ratio. ABSTRACT: Recently, breast types were categorized into four types based on the Breast Imaging Reporting and Data System (BI-RADS) atlas, and evaluating them is vital in clinical practice. A Japanese guideline, called breast composition, was developed for the breast types based on BI-RADS. The guideline is characterized using a continuous value called the mammary gland content ratio calculated to determine the breast composition, therefore allowing a more objective and visual evaluation. Although a discriminative deep convolutional neural network (DCNN) has been developed conventionally to classify the breast composition, it could encounter two-step errors or more. Hence, we propose an alternative regression DCNN based on mammary gland content ratio. We used 1476 images, evaluated by an expert physician. Our regression DCNN contained four convolution layers and three fully connected layers. Consequently, we obtained a high correlation of 0.93 (p < 0.01). Furthermore, to scrutinize the effectiveness of the regression DCNN, we categorized breast composition using the estimated ratio obtained by the regression DCNN. The agreement rates are high at 84.8%, suggesting that the breast composition can be calculated using regression DCNN with high accuracy. Moreover, the occurrence of two-step errors or more is unlikely, and the proposed method can intuitively understand the estimated results. |
format | Online Article Text |
id | pubmed-10216583 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2023 |
publisher | MDPI |
record_format | MEDLINE/PubMed |
spelling | pubmed-102165832023-05-27 Automated Estimation of Mammary Gland Content Ratio Using Regression Deep Convolutional Neural Network and the Effectiveness in Clinical Practice as Explainable Artificial Intelligence Kai, Chiharu Ishizuka, Sachi Otsuka, Tsunehiro Nara, Miyako Kondo, Satoshi Futamura, Hitoshi Kodama, Naoki Kasai, Satoshi Cancers (Basel) Article SIMPLE SUMMARY: In Japan, a guideline for breast types, called “breast composition,” was recently developed based on BI-RADS. The Japanese guidelines are characterized using a continuous value called the mammary gland content ratio, calculated using the density of the pectoralis muscle as an indicator to determine breast composition. Discriminative DCNN has been developed conventionally to classify breast composition; however, it could encounter two-step errors or more (e.g., estimating “Fatty” as “Heterogeneous dense”). We developed a regression DCNN based on the mammary gland content ratio defined in the Japanese guideline to solve the above problem, followed by automated breast composition classification based on the continuous value. We also examined the usefulness of the continuous value of the mammary gland content ratio. ABSTRACT: Recently, breast types were categorized into four types based on the Breast Imaging Reporting and Data System (BI-RADS) atlas, and evaluating them is vital in clinical practice. A Japanese guideline, called breast composition, was developed for the breast types based on BI-RADS. The guideline is characterized using a continuous value called the mammary gland content ratio calculated to determine the breast composition, therefore allowing a more objective and visual evaluation. Although a discriminative deep convolutional neural network (DCNN) has been developed conventionally to classify the breast composition, it could encounter two-step errors or more. Hence, we propose an alternative regression DCNN based on mammary gland content ratio. We used 1476 images, evaluated by an expert physician. Our regression DCNN contained four convolution layers and three fully connected layers. Consequently, we obtained a high correlation of 0.93 (p < 0.01). Furthermore, to scrutinize the effectiveness of the regression DCNN, we categorized breast composition using the estimated ratio obtained by the regression DCNN. The agreement rates are high at 84.8%, suggesting that the breast composition can be calculated using regression DCNN with high accuracy. Moreover, the occurrence of two-step errors or more is unlikely, and the proposed method can intuitively understand the estimated results. MDPI 2023-05-17 /pmc/articles/PMC10216583/ /pubmed/37345132 http://dx.doi.org/10.3390/cancers15102794 Text en © 2023 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). |
spellingShingle | Article Kai, Chiharu Ishizuka, Sachi Otsuka, Tsunehiro Nara, Miyako Kondo, Satoshi Futamura, Hitoshi Kodama, Naoki Kasai, Satoshi Automated Estimation of Mammary Gland Content Ratio Using Regression Deep Convolutional Neural Network and the Effectiveness in Clinical Practice as Explainable Artificial Intelligence |
title | Automated Estimation of Mammary Gland Content Ratio Using Regression Deep Convolutional Neural Network and the Effectiveness in Clinical Practice as Explainable Artificial Intelligence |
title_full | Automated Estimation of Mammary Gland Content Ratio Using Regression Deep Convolutional Neural Network and the Effectiveness in Clinical Practice as Explainable Artificial Intelligence |
title_fullStr | Automated Estimation of Mammary Gland Content Ratio Using Regression Deep Convolutional Neural Network and the Effectiveness in Clinical Practice as Explainable Artificial Intelligence |
title_full_unstemmed | Automated Estimation of Mammary Gland Content Ratio Using Regression Deep Convolutional Neural Network and the Effectiveness in Clinical Practice as Explainable Artificial Intelligence |
title_short | Automated Estimation of Mammary Gland Content Ratio Using Regression Deep Convolutional Neural Network and the Effectiveness in Clinical Practice as Explainable Artificial Intelligence |
title_sort | automated estimation of mammary gland content ratio using regression deep convolutional neural network and the effectiveness in clinical practice as explainable artificial intelligence |
topic | Article |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10216583/ https://www.ncbi.nlm.nih.gov/pubmed/37345132 http://dx.doi.org/10.3390/cancers15102794 |
work_keys_str_mv | AT kaichiharu automatedestimationofmammaryglandcontentratiousingregressiondeepconvolutionalneuralnetworkandtheeffectivenessinclinicalpracticeasexplainableartificialintelligence AT ishizukasachi automatedestimationofmammaryglandcontentratiousingregressiondeepconvolutionalneuralnetworkandtheeffectivenessinclinicalpracticeasexplainableartificialintelligence AT otsukatsunehiro automatedestimationofmammaryglandcontentratiousingregressiondeepconvolutionalneuralnetworkandtheeffectivenessinclinicalpracticeasexplainableartificialintelligence AT naramiyako automatedestimationofmammaryglandcontentratiousingregressiondeepconvolutionalneuralnetworkandtheeffectivenessinclinicalpracticeasexplainableartificialintelligence AT kondosatoshi automatedestimationofmammaryglandcontentratiousingregressiondeepconvolutionalneuralnetworkandtheeffectivenessinclinicalpracticeasexplainableartificialintelligence AT futamurahitoshi automatedestimationofmammaryglandcontentratiousingregressiondeepconvolutionalneuralnetworkandtheeffectivenessinclinicalpracticeasexplainableartificialintelligence AT kodamanaoki automatedestimationofmammaryglandcontentratiousingregressiondeepconvolutionalneuralnetworkandtheeffectivenessinclinicalpracticeasexplainableartificialintelligence AT kasaisatoshi automatedestimationofmammaryglandcontentratiousingregressiondeepconvolutionalneuralnetworkandtheeffectivenessinclinicalpracticeasexplainableartificialintelligence |