Data Augmentation Using Adversarial Image-to-Image Translation for the Segmentation of Mobile-Acquired Dermatological Images
Dermoscopic images allow the detailed examination of subsurface characteristics of the skin, which led to creating several substantial databases of diverse skin lesions. However, the dermoscope is not an easily accessible tool in some regions. A less expensive alternative could be acquiring medium r...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-12-01
|
| Series: | Journal of Imaging |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2313-433X/7/1/2 |
| _version_ | 1797543599119794176 |
|---|---|
| author | Catarina Andrade Luís F. Teixeira Maria João M. Vasconcelos Luís Rosado |
| author_facet | Catarina Andrade Luís F. Teixeira Maria João M. Vasconcelos Luís Rosado |
| author_sort | Catarina Andrade |
| collection | DOAJ |
| description | Dermoscopic images allow the detailed examination of subsurface characteristics of the skin, which led to creating several substantial databases of diverse skin lesions. However, the dermoscope is not an easily accessible tool in some regions. A less expensive alternative could be acquiring medium resolution clinical macroscopic images of skin lesions. However, the limited volume of macroscopic images available, especially mobile-acquired, hinders developing a clinical mobile-based deep learning approach. In this work, we present a technique to efficiently utilize the sizable number of dermoscopic images to improve the segmentation capacity of macroscopic skin lesion images. A Cycle-Consistent Adversarial Network is used to translate the image between the two distinct domains created by the different image acquisition devices. A visual inspection was performed on several databases for qualitative evaluation of the results, based on the disappearance and appearance of intrinsic dermoscopic and macroscopic features. Moreover, the Fréchet Inception Distance was used as a quantitative metric. The quantitative segmentation results are demonstrated on the available macroscopic segmentation databases, SMARTSKINS and Dermofit Image Library, yielding test set thresholded Jaccard Index of 85.13% and 74.30%. These results establish a new state-of-the-art performance in the SMARTSKINS database. |
| first_indexed | 2024-03-10T13:47:56Z |
| format | Article |
| id | doaj.art-e86ad7c73a2b48b69bf700b20f38c578 |
| institution | Directory Open Access Journal |
| issn | 2313-433X |
| language | English |
| last_indexed | 2024-03-10T13:47:56Z |
| publishDate | 2020-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Imaging |
| spelling | doaj.art-e86ad7c73a2b48b69bf700b20f38c5782023-11-21T02:27:01ZengMDPI AGJournal of Imaging2313-433X2020-12-0171210.3390/jimaging7010002Data Augmentation Using Adversarial Image-to-Image Translation for the Segmentation of Mobile-Acquired Dermatological ImagesCatarina Andrade0Luís F. Teixeira1Maria João M. Vasconcelos2Luís Rosado3Fraunhofer Portugal AICOS, Rua Alfredo Allen, 4200-135 Porto, PortugalFaculty of Engineering, University of Porto, Rua Dr. Roberto Frias, 4200-465 Porto, PortugalFraunhofer Portugal AICOS, Rua Alfredo Allen, 4200-135 Porto, PortugalFraunhofer Portugal AICOS, Rua Alfredo Allen, 4200-135 Porto, PortugalDermoscopic images allow the detailed examination of subsurface characteristics of the skin, which led to creating several substantial databases of diverse skin lesions. However, the dermoscope is not an easily accessible tool in some regions. A less expensive alternative could be acquiring medium resolution clinical macroscopic images of skin lesions. However, the limited volume of macroscopic images available, especially mobile-acquired, hinders developing a clinical mobile-based deep learning approach. In this work, we present a technique to efficiently utilize the sizable number of dermoscopic images to improve the segmentation capacity of macroscopic skin lesion images. A Cycle-Consistent Adversarial Network is used to translate the image between the two distinct domains created by the different image acquisition devices. A visual inspection was performed on several databases for qualitative evaluation of the results, based on the disappearance and appearance of intrinsic dermoscopic and macroscopic features. Moreover, the Fréchet Inception Distance was used as a quantitative metric. The quantitative segmentation results are demonstrated on the available macroscopic segmentation databases, SMARTSKINS and Dermofit Image Library, yielding test set thresholded Jaccard Index of 85.13% and 74.30%. These results establish a new state-of-the-art performance in the SMARTSKINS database.https://www.mdpi.com/2313-433X/7/1/2convolutional neural networkCycleGANdata augmentationdermoscopic imagesdomain transfermacroscopic images |
| spellingShingle | Catarina Andrade Luís F. Teixeira Maria João M. Vasconcelos Luís Rosado Data Augmentation Using Adversarial Image-to-Image Translation for the Segmentation of Mobile-Acquired Dermatological Images Journal of Imaging convolutional neural network CycleGAN data augmentation dermoscopic images domain transfer macroscopic images |
| title | Data Augmentation Using Adversarial Image-to-Image Translation for the Segmentation of Mobile-Acquired Dermatological Images |
| title_full | Data Augmentation Using Adversarial Image-to-Image Translation for the Segmentation of Mobile-Acquired Dermatological Images |
| title_fullStr | Data Augmentation Using Adversarial Image-to-Image Translation for the Segmentation of Mobile-Acquired Dermatological Images |
| title_full_unstemmed | Data Augmentation Using Adversarial Image-to-Image Translation for the Segmentation of Mobile-Acquired Dermatological Images |
| title_short | Data Augmentation Using Adversarial Image-to-Image Translation for the Segmentation of Mobile-Acquired Dermatological Images |
| title_sort | data augmentation using adversarial image to image translation for the segmentation of mobile acquired dermatological images |
| topic | convolutional neural network CycleGAN data augmentation dermoscopic images domain transfer macroscopic images |
| url | https://www.mdpi.com/2313-433X/7/1/2 |
| work_keys_str_mv | AT catarinaandrade dataaugmentationusingadversarialimagetoimagetranslationforthesegmentationofmobileacquireddermatologicalimages AT luisfteixeira dataaugmentationusingadversarialimagetoimagetranslationforthesegmentationofmobileacquireddermatologicalimages AT mariajoaomvasconcelos dataaugmentationusingadversarialimagetoimagetranslationforthesegmentationofmobileacquireddermatologicalimages AT luisrosado dataaugmentationusingadversarialimagetoimagetranslationforthesegmentationofmobileacquireddermatologicalimages |