Eintrag weiter verarbeiten
Fully automatic classification of breast MRI background parenchymal enhancement using a transfer learning approach
Gespeichert in:
| Zeitschriftentitel: | Medicine |
|---|---|
| Personen und Körperschaften: | , , , , , , , |
| In: | Medicine, 99, 2020, 29, S. e21243 |
| Format: | E-Article |
| Sprache: | Englisch |
| veröffentlicht: |
Ovid Technologies (Wolters Kluwer Health)
|
| Schlagwörter: |
| author_facet |
Borkowski, Karol Rossi, Cristina Ciritsis, Alexander Marcon, Magda Hejduk, Patryk Stieb, Sonja Boss, Andreas Berger, Nicole Borkowski, Karol Rossi, Cristina Ciritsis, Alexander Marcon, Magda Hejduk, Patryk Stieb, Sonja Boss, Andreas Berger, Nicole |
|---|---|
| author |
Borkowski, Karol Rossi, Cristina Ciritsis, Alexander Marcon, Magda Hejduk, Patryk Stieb, Sonja Boss, Andreas Berger, Nicole |
| spellingShingle |
Borkowski, Karol Rossi, Cristina Ciritsis, Alexander Marcon, Magda Hejduk, Patryk Stieb, Sonja Boss, Andreas Berger, Nicole Medicine Fully automatic classification of breast MRI background parenchymal enhancement using a transfer learning approach General Medicine |
| author_sort |
borkowski, karol |
| spelling |
Borkowski, Karol Rossi, Cristina Ciritsis, Alexander Marcon, Magda Hejduk, Patryk Stieb, Sonja Boss, Andreas Berger, Nicole 0025-7974 1536-5964 Ovid Technologies (Wolters Kluwer Health) General Medicine http://dx.doi.org/10.1097/md.0000000000021243 <jats:sec> <jats:title>Abstract</jats:title> <jats:p>Marked enhancement of the fibroglandular tissue on contrast-enhanced breast magnetic resonance imaging (MRI) may affect lesion detection and classification and is suggested to be associated with higher risk of developing breast cancer. The background parenchymal enhancement (BPE) is qualitatively classified according to the BI-RADS atlas into the categories “minimal,” “mild,” “moderate,” and “marked.” The purpose of this study was to train a deep convolutional neural network (dCNN) for standardized and automatic classification of BPE categories.</jats:p> <jats:p>This IRB-approved retrospective study included 11,769 single MR images from 149 patients. The MR images were derived from the subtraction between the first post-contrast volume and the native T1-weighted images. A hierarchic approach was implemented relying on 2 dCNN models for detection of MR-slices imaging breast tissue and for BPE classification, respectively. Data annotation was performed by 2 board-certified radiologists. The consensus of the 2 radiologists was chosen as reference for BPE classification. The clinical performances of the single readers and of the dCNN were statistically compared using the quadratic Cohen's kappa.</jats:p> <jats:p>Slices depicting the breast were classified with training, validation, and real-world (test) accuracies of 98%, 96%, and 97%, respectively. Over the 4 classes, the BPE classification was reached with mean accuracies of 74% for training, 75% for the validation, and 75% for the real word dataset. As compared to the reference, the inter-reader reliabilities for the radiologists were 0.780 (reader 1) and 0.679 (reader 2). On the other hand, the reliability for the dCNN model was 0.815.</jats:p> <jats:p>Automatic classification of BPE can be performed with high accuracy and support the standardization of tissue classification in MRI.</jats:p> </jats:sec> Fully automatic classification of breast MRI background parenchymal enhancement using a transfer learning approach Medicine |
| doi_str_mv |
10.1097/md.0000000000021243 |
| facet_avail |
Online Free |
| format |
ElectronicArticle |
| fullrecord |
blob:ai-49-aHR0cDovL2R4LmRvaS5vcmcvMTAuMTA5Ny9tZC4wMDAwMDAwMDAwMDIxMjQz |
| id |
ai-49-aHR0cDovL2R4LmRvaS5vcmcvMTAuMTA5Ny9tZC4wMDAwMDAwMDAwMDIxMjQz |
| institution |
DE-15 DE-Pl11 DE-Rs1 DE-105 DE-14 DE-Ch1 DE-L229 DE-D275 DE-Bn3 DE-Brt1 DE-D161 DE-Zwi2 DE-Gla1 DE-Zi4 |
| imprint |
Ovid Technologies (Wolters Kluwer Health), 2020 |
| imprint_str_mv |
Ovid Technologies (Wolters Kluwer Health), 2020 |
| issn |
0025-7974 1536-5964 |
| issn_str_mv |
0025-7974 1536-5964 |
| language |
English |
| mega_collection |
Ovid Technologies (Wolters Kluwer Health) (CrossRef) |
| match_str |
borkowski2020fullyautomaticclassificationofbreastmribackgroundparenchymalenhancementusingatransferlearningapproach |
| publishDateSort |
2020 |
| publisher |
Ovid Technologies (Wolters Kluwer Health) |
| recordtype |
ai |
| record_format |
ai |
| series |
Medicine |
| source_id |
49 |
| title |
Fully automatic classification of breast MRI background parenchymal enhancement using a transfer learning approach |
| title_unstemmed |
Fully automatic classification of breast MRI background parenchymal enhancement using a transfer learning approach |
| title_full |
Fully automatic classification of breast MRI background parenchymal enhancement using a transfer learning approach |
| title_fullStr |
Fully automatic classification of breast MRI background parenchymal enhancement using a transfer learning approach |
| title_full_unstemmed |
Fully automatic classification of breast MRI background parenchymal enhancement using a transfer learning approach |
| title_short |
Fully automatic classification of breast MRI background parenchymal enhancement using a transfer learning approach |
| title_sort |
fully automatic classification of breast mri background parenchymal enhancement using a transfer learning approach |
| topic |
General Medicine |
| url |
http://dx.doi.org/10.1097/md.0000000000021243 |
| publishDate |
2020 |
| physical |
e21243 |
| description |
<jats:sec>
<jats:title>Abstract</jats:title>
<jats:p>Marked enhancement of the fibroglandular tissue on contrast-enhanced breast magnetic resonance imaging (MRI) may affect lesion detection and classification and is suggested to be associated with higher risk of developing breast cancer. The background parenchymal enhancement (BPE) is qualitatively classified according to the BI-RADS atlas into the categories “minimal,” “mild,” “moderate,” and “marked.” The purpose of this study was to train a deep convolutional neural network (dCNN) for standardized and automatic classification of BPE categories.</jats:p>
<jats:p>This IRB-approved retrospective study included 11,769 single MR images from 149 patients. The MR images were derived from the subtraction between the first post-contrast volume and the native T1-weighted images. A hierarchic approach was implemented relying on 2 dCNN models for detection of MR-slices imaging breast tissue and for BPE classification, respectively. Data annotation was performed by 2 board-certified radiologists. The consensus of the 2 radiologists was chosen as reference for BPE classification. The clinical performances of the single readers and of the dCNN were statistically compared using the quadratic Cohen's kappa.</jats:p>
<jats:p>Slices depicting the breast were classified with training, validation, and real-world (test) accuracies of 98%, 96%, and 97%, respectively. Over the 4 classes, the BPE classification was reached with mean accuracies of 74% for training, 75% for the validation, and 75% for the real word dataset. As compared to the reference, the inter-reader reliabilities for the radiologists were 0.780 (reader 1) and 0.679 (reader 2). On the other hand, the reliability for the dCNN model was 0.815.</jats:p>
<jats:p>Automatic classification of BPE can be performed with high accuracy and support the standardization of tissue classification in MRI.</jats:p>
</jats:sec> |
| container_issue |
29 |
| container_start_page |
0 |
| container_title |
Medicine |
| container_volume |
99 |
| format_de105 |
Article, E-Article |
| format_de14 |
Article, E-Article |
| format_de15 |
Article, E-Article |
| format_de520 |
Article, E-Article |
| format_de540 |
Article, E-Article |
| format_dech1 |
Article, E-Article |
| format_ded117 |
Article, E-Article |
| format_degla1 |
E-Article |
| format_del152 |
Buch |
| format_del189 |
Article, E-Article |
| format_dezi4 |
Article |
| format_dezwi2 |
Article, E-Article |
| format_finc |
Article, E-Article |
| format_nrw |
Article, E-Article |
| _version_ |
1792346165381955585 |
| geogr_code |
not assigned |
| last_indexed |
2024-03-01T17:35:01.533Z |
| geogr_code_person |
not assigned |
| openURL |
url_ver=Z39.88-2004&ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fvufind.svn.sourceforge.net%3Agenerator&rft.title=Fully+automatic+classification+of+breast+MRI+background+parenchymal+enhancement+using+a+transfer+learning+approach&rft.date=2020-07-17&genre=article&issn=1536-5964&volume=99&issue=29&pages=e21243&jtitle=Medicine&atitle=Fully+automatic+classification+of+breast+MRI+background+parenchymal+enhancement+using+a+transfer+learning+approach&aulast=Berger&aufirst=Nicole&rft_id=info%3Adoi%2F10.1097%2Fmd.0000000000021243&rft.language%5B0%5D=eng |
| SOLR | |
| _version_ | 1792346165381955585 |
| author | Borkowski, Karol, Rossi, Cristina, Ciritsis, Alexander, Marcon, Magda, Hejduk, Patryk, Stieb, Sonja, Boss, Andreas, Berger, Nicole |
| author_facet | Borkowski, Karol, Rossi, Cristina, Ciritsis, Alexander, Marcon, Magda, Hejduk, Patryk, Stieb, Sonja, Boss, Andreas, Berger, Nicole, Borkowski, Karol, Rossi, Cristina, Ciritsis, Alexander, Marcon, Magda, Hejduk, Patryk, Stieb, Sonja, Boss, Andreas, Berger, Nicole |
| author_sort | borkowski, karol |
| container_issue | 29 |
| container_start_page | 0 |
| container_title | Medicine |
| container_volume | 99 |
| description | <jats:sec> <jats:title>Abstract</jats:title> <jats:p>Marked enhancement of the fibroglandular tissue on contrast-enhanced breast magnetic resonance imaging (MRI) may affect lesion detection and classification and is suggested to be associated with higher risk of developing breast cancer. The background parenchymal enhancement (BPE) is qualitatively classified according to the BI-RADS atlas into the categories “minimal,” “mild,” “moderate,” and “marked.” The purpose of this study was to train a deep convolutional neural network (dCNN) for standardized and automatic classification of BPE categories.</jats:p> <jats:p>This IRB-approved retrospective study included 11,769 single MR images from 149 patients. The MR images were derived from the subtraction between the first post-contrast volume and the native T1-weighted images. A hierarchic approach was implemented relying on 2 dCNN models for detection of MR-slices imaging breast tissue and for BPE classification, respectively. Data annotation was performed by 2 board-certified radiologists. The consensus of the 2 radiologists was chosen as reference for BPE classification. The clinical performances of the single readers and of the dCNN were statistically compared using the quadratic Cohen's kappa.</jats:p> <jats:p>Slices depicting the breast were classified with training, validation, and real-world (test) accuracies of 98%, 96%, and 97%, respectively. Over the 4 classes, the BPE classification was reached with mean accuracies of 74% for training, 75% for the validation, and 75% for the real word dataset. As compared to the reference, the inter-reader reliabilities for the radiologists were 0.780 (reader 1) and 0.679 (reader 2). On the other hand, the reliability for the dCNN model was 0.815.</jats:p> <jats:p>Automatic classification of BPE can be performed with high accuracy and support the standardization of tissue classification in MRI.</jats:p> </jats:sec> |
| doi_str_mv | 10.1097/md.0000000000021243 |
| facet_avail | Online, Free |
| format | ElectronicArticle |
| format_de105 | Article, E-Article |
| format_de14 | Article, E-Article |
| format_de15 | Article, E-Article |
| format_de520 | Article, E-Article |
| format_de540 | Article, E-Article |
| format_dech1 | Article, E-Article |
| format_ded117 | Article, E-Article |
| format_degla1 | E-Article |
| format_del152 | Buch |
| format_del189 | Article, E-Article |
| format_dezi4 | Article |
| format_dezwi2 | Article, E-Article |
| format_finc | Article, E-Article |
| format_nrw | Article, E-Article |
| geogr_code | not assigned |
| geogr_code_person | not assigned |
| id | ai-49-aHR0cDovL2R4LmRvaS5vcmcvMTAuMTA5Ny9tZC4wMDAwMDAwMDAwMDIxMjQz |
| imprint | Ovid Technologies (Wolters Kluwer Health), 2020 |
| imprint_str_mv | Ovid Technologies (Wolters Kluwer Health), 2020 |
| institution | DE-15, DE-Pl11, DE-Rs1, DE-105, DE-14, DE-Ch1, DE-L229, DE-D275, DE-Bn3, DE-Brt1, DE-D161, DE-Zwi2, DE-Gla1, DE-Zi4 |
| issn | 0025-7974, 1536-5964 |
| issn_str_mv | 0025-7974, 1536-5964 |
| language | English |
| last_indexed | 2024-03-01T17:35:01.533Z |
| match_str | borkowski2020fullyautomaticclassificationofbreastmribackgroundparenchymalenhancementusingatransferlearningapproach |
| mega_collection | Ovid Technologies (Wolters Kluwer Health) (CrossRef) |
| physical | e21243 |
| publishDate | 2020 |
| publishDateSort | 2020 |
| publisher | Ovid Technologies (Wolters Kluwer Health) |
| record_format | ai |
| recordtype | ai |
| series | Medicine |
| source_id | 49 |
| spelling | Borkowski, Karol Rossi, Cristina Ciritsis, Alexander Marcon, Magda Hejduk, Patryk Stieb, Sonja Boss, Andreas Berger, Nicole 0025-7974 1536-5964 Ovid Technologies (Wolters Kluwer Health) General Medicine http://dx.doi.org/10.1097/md.0000000000021243 <jats:sec> <jats:title>Abstract</jats:title> <jats:p>Marked enhancement of the fibroglandular tissue on contrast-enhanced breast magnetic resonance imaging (MRI) may affect lesion detection and classification and is suggested to be associated with higher risk of developing breast cancer. The background parenchymal enhancement (BPE) is qualitatively classified according to the BI-RADS atlas into the categories “minimal,” “mild,” “moderate,” and “marked.” The purpose of this study was to train a deep convolutional neural network (dCNN) for standardized and automatic classification of BPE categories.</jats:p> <jats:p>This IRB-approved retrospective study included 11,769 single MR images from 149 patients. The MR images were derived from the subtraction between the first post-contrast volume and the native T1-weighted images. A hierarchic approach was implemented relying on 2 dCNN models for detection of MR-slices imaging breast tissue and for BPE classification, respectively. Data annotation was performed by 2 board-certified radiologists. The consensus of the 2 radiologists was chosen as reference for BPE classification. The clinical performances of the single readers and of the dCNN were statistically compared using the quadratic Cohen's kappa.</jats:p> <jats:p>Slices depicting the breast were classified with training, validation, and real-world (test) accuracies of 98%, 96%, and 97%, respectively. Over the 4 classes, the BPE classification was reached with mean accuracies of 74% for training, 75% for the validation, and 75% for the real word dataset. As compared to the reference, the inter-reader reliabilities for the radiologists were 0.780 (reader 1) and 0.679 (reader 2). On the other hand, the reliability for the dCNN model was 0.815.</jats:p> <jats:p>Automatic classification of BPE can be performed with high accuracy and support the standardization of tissue classification in MRI.</jats:p> </jats:sec> Fully automatic classification of breast MRI background parenchymal enhancement using a transfer learning approach Medicine |
| spellingShingle | Borkowski, Karol, Rossi, Cristina, Ciritsis, Alexander, Marcon, Magda, Hejduk, Patryk, Stieb, Sonja, Boss, Andreas, Berger, Nicole, Medicine, Fully automatic classification of breast MRI background parenchymal enhancement using a transfer learning approach, General Medicine |
| title | Fully automatic classification of breast MRI background parenchymal enhancement using a transfer learning approach |
| title_full | Fully automatic classification of breast MRI background parenchymal enhancement using a transfer learning approach |
| title_fullStr | Fully automatic classification of breast MRI background parenchymal enhancement using a transfer learning approach |
| title_full_unstemmed | Fully automatic classification of breast MRI background parenchymal enhancement using a transfer learning approach |
| title_short | Fully automatic classification of breast MRI background parenchymal enhancement using a transfer learning approach |
| title_sort | fully automatic classification of breast mri background parenchymal enhancement using a transfer learning approach |
| title_unstemmed | Fully automatic classification of breast MRI background parenchymal enhancement using a transfer learning approach |
| topic | General Medicine |
| url | http://dx.doi.org/10.1097/md.0000000000021243 |