Generative-Adversarial-Network-Based Data Augmentation for the Classification of Craniosynostosis
Craniosynostosis is a congenital disease characterized by the premature closure of one or multiple sutures of the infant’s skull. For diagnosis, 3D photogrammetric scans are a radiation-free alternative to computed tomography. However, data is only sparsely available and the role of data augmentatio...
Main Authors: | , , , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
De Gruyter
2022-09-01
|
Series: | Current Directions in Biomedical Engineering |
Subjects: | |
Online Access: | https://doi.org/10.1515/cdbme-2022-1005 |
_version_ | 1797947878847545344 |
---|---|
author | Kaiser Christian Schaufelberger Matthias Kühle Reinald Peter Wachter Andreas Weichel Frederic Hagen Niclas Ringwald Friedemann Eisenmann Urs Engel Michael Freudlsperger Christian Nahm Werner |
author_facet | Kaiser Christian Schaufelberger Matthias Kühle Reinald Peter Wachter Andreas Weichel Frederic Hagen Niclas Ringwald Friedemann Eisenmann Urs Engel Michael Freudlsperger Christian Nahm Werner |
author_sort | Kaiser Christian |
collection | DOAJ |
description | Craniosynostosis is a congenital disease characterized by the premature closure of one or multiple sutures of the infant’s skull. For diagnosis, 3D photogrammetric scans are a radiation-free alternative to computed tomography. However, data is only sparsely available and the role of data augmentation for the classification of craniosynostosis has not yet been analyzed. In this work, we use a 2D distance map representation of the infants’ heads with a convolutional-neural-network-based classifier and employ a generative adversarial network (GAN) for data augmentation. We simulate two data scarcity scenarios with 15% and 10% training data and test the influence of different degrees of added synthetic data and balancing underrepresented classes. We used total accuracy and F1-score as a metric to evaluate the final classifiers. For 15% training data, the GAN-augmented dataset showed an increased F1-score up to 0.1 and classification accuracy up to 3 %. For 10% training data, both metrics decreased. We present a deep convolutional GAN capable of creating synthetic data for the classification of craniosynostosis. Using a moderate amount of synthetic data using a GAN showed slightly better performance, but had little effect overall. The simulated scarcity scenario of 10% training data may have limited the model’s ability to learn the underlying data distribution. |
first_indexed | 2024-04-10T21:33:34Z |
format | Article |
id | doaj.art-3e928e73d32845a2848207d68256b935 |
institution | Directory Open Access Journal |
issn | 2364-5504 |
language | English |
last_indexed | 2024-04-10T21:33:34Z |
publishDate | 2022-09-01 |
publisher | De Gruyter |
record_format | Article |
series | Current Directions in Biomedical Engineering |
spelling | doaj.art-3e928e73d32845a2848207d68256b9352023-01-19T12:47:02ZengDe GruyterCurrent Directions in Biomedical Engineering2364-55042022-09-0182172010.1515/cdbme-2022-1005Generative-Adversarial-Network-Based Data Augmentation for the Classification of CraniosynostosisKaiser Christian0Schaufelberger Matthias1Kühle Reinald Peter2Wachter Andreas3Weichel Frederic4Hagen Niclas5Ringwald Friedemann6Eisenmann Urs7Engel Michael8Freudlsperger Christian9Nahm Werner10Institute of Biomedical Engineering (IBT), Karlsruhe Institute of Technology (KIT), Kaiserstr. 12,Karlsruhe, GermanyInstitute of Biomedical Engineering (IBT), Karlsruhe Institute of Technology (KIT), Kaiserstr. 12,Karlsruhe, GermanyDepartment of Oral and Maxillofacial Surgery, Heidelberg University Hospital, Im Neuenheimer Feld 400,Heidelberg, GermanyInstitute of Biomedical Engineering (IBT), Karlsruhe Institute of Technology (KIT), Kaiserstr. 12,Karlsruhe, GermanyDepartment of Oral and Maxillofacial Surgery, Heidelberg University Hospital, Im Neuenheimer Feld 400,Heidelberg, GermanyInstitute of Medical Informatics, Heidelberg University Hospital, Im Neuenheimer Feld 130.3,Heidelberg, GermanyInstitute of Medical Informatics, Heidelberg University Hospital, Im Neuenheimer Feld 130.3,Heidelberg, GermanyInstitute of Medical Informatics, Heidelberg University Hospital, Im Neuenheimer Feld 130.3,Heidelberg, GermanyDepartment of Oral and Maxillofacial Surgery, Heidelberg University Hospital, Im Neuenheimer Feld 400,Heidelberg, GermanyDepartment of Oral and Maxillofacial Surgery, Heidelberg University Hospital, Im Neuenheimer Feld 400,Heidelberg, GermanyInstitute of Biomedical Engineering (IBT), Karlsruhe Institute of Technology (KIT), Kaiserstr. 12,Karlsruhe, GermanyCraniosynostosis is a congenital disease characterized by the premature closure of one or multiple sutures of the infant’s skull. For diagnosis, 3D photogrammetric scans are a radiation-free alternative to computed tomography. However, data is only sparsely available and the role of data augmentation for the classification of craniosynostosis has not yet been analyzed. In this work, we use a 2D distance map representation of the infants’ heads with a convolutional-neural-network-based classifier and employ a generative adversarial network (GAN) for data augmentation. We simulate two data scarcity scenarios with 15% and 10% training data and test the influence of different degrees of added synthetic data and balancing underrepresented classes. We used total accuracy and F1-score as a metric to evaluate the final classifiers. For 15% training data, the GAN-augmented dataset showed an increased F1-score up to 0.1 and classification accuracy up to 3 %. For 10% training data, both metrics decreased. We present a deep convolutional GAN capable of creating synthetic data for the classification of craniosynostosis. Using a moderate amount of synthetic data using a GAN showed slightly better performance, but had little effect overall. The simulated scarcity scenario of 10% training data may have limited the model’s ability to learn the underlying data distribution.https://doi.org/10.1515/cdbme-2022-1005generative adversarial networkclassificationcraniosynostosisdata augmentation |
spellingShingle | Kaiser Christian Schaufelberger Matthias Kühle Reinald Peter Wachter Andreas Weichel Frederic Hagen Niclas Ringwald Friedemann Eisenmann Urs Engel Michael Freudlsperger Christian Nahm Werner Generative-Adversarial-Network-Based Data Augmentation for the Classification of Craniosynostosis Current Directions in Biomedical Engineering generative adversarial network classification craniosynostosis data augmentation |
title | Generative-Adversarial-Network-Based Data Augmentation for the Classification of Craniosynostosis |
title_full | Generative-Adversarial-Network-Based Data Augmentation for the Classification of Craniosynostosis |
title_fullStr | Generative-Adversarial-Network-Based Data Augmentation for the Classification of Craniosynostosis |
title_full_unstemmed | Generative-Adversarial-Network-Based Data Augmentation for the Classification of Craniosynostosis |
title_short | Generative-Adversarial-Network-Based Data Augmentation for the Classification of Craniosynostosis |
title_sort | generative adversarial network based data augmentation for the classification of craniosynostosis |
topic | generative adversarial network classification craniosynostosis data augmentation |
url | https://doi.org/10.1515/cdbme-2022-1005 |
work_keys_str_mv | AT kaiserchristian generativeadversarialnetworkbaseddataaugmentationfortheclassificationofcraniosynostosis AT schaufelbergermatthias generativeadversarialnetworkbaseddataaugmentationfortheclassificationofcraniosynostosis AT kuhlereinaldpeter generativeadversarialnetworkbaseddataaugmentationfortheclassificationofcraniosynostosis AT wachterandreas generativeadversarialnetworkbaseddataaugmentationfortheclassificationofcraniosynostosis AT weichelfrederic generativeadversarialnetworkbaseddataaugmentationfortheclassificationofcraniosynostosis AT hagenniclas generativeadversarialnetworkbaseddataaugmentationfortheclassificationofcraniosynostosis AT ringwaldfriedemann generativeadversarialnetworkbaseddataaugmentationfortheclassificationofcraniosynostosis AT eisenmannurs generativeadversarialnetworkbaseddataaugmentationfortheclassificationofcraniosynostosis AT engelmichael generativeadversarialnetworkbaseddataaugmentationfortheclassificationofcraniosynostosis AT freudlspergerchristian generativeadversarialnetworkbaseddataaugmentationfortheclassificationofcraniosynostosis AT nahmwerner generativeadversarialnetworkbaseddataaugmentationfortheclassificationofcraniosynostosis |