Skin Lesion Segmentation in Dermoscopic Images With Ensemble Deep Learning Methods
Early detection of skin cancer, particularly melanoma, is crucial to enable advanced treatment. Due to the rapid growth in the number of skin cancers, there is a growing need of computerised analysis for skin lesions. The state-of-the-art public available datasets for skin lesions are often accompan...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2020-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/8936444/ |
| _version_ | 1831806574794899456 |
|---|---|
| author | Manu Goyal Amanda Oakley Priyanka Bansal Darren Dancey Moi Hoon Yap |
| author_facet | Manu Goyal Amanda Oakley Priyanka Bansal Darren Dancey Moi Hoon Yap |
| author_sort | Manu Goyal |
| collection | DOAJ |
| description | Early detection of skin cancer, particularly melanoma, is crucial to enable advanced treatment. Due to the rapid growth in the number of skin cancers, there is a growing need of computerised analysis for skin lesions. The state-of-the-art public available datasets for skin lesions are often accompanied with a very limited amount of segmentation ground truth labeling. Also, the available segmentation datasets consist of noisy expert annotations reflecting the fact that precise annotations to represent the boundary of skin lesions are laborious and expensive. The lesion boundary segmentation is vital to locate the lesion accurately in dermoscopic images and lesion diagnosis of different skin lesion types. In this work, we propose the fully automated deep learning ensemble methods to achieve high sensitivity and high specificity in lesion boundary segmentation. We trained the ensemble methods based on Mask R-CNN and DeeplabV3+ methods on ISIC-2017 segmentation training set and evaluate the performance of the ensemble networks on ISIC-2017 testing set and PH2 dataset. Our results showed that the proposed ensemble methods segmented the skin lesions with Sensitivity of 89.93% and Specificity of 97.94% for the ISIC-2017 testing set. The proposed ensemble method Ensemble-A outperformed FrCN, FCNs, U-Net, and SegNet in Sensitivity by 4.4%, 8.8%, 22.7%, and 9.8% respectively. Furthermore, the proposed ensemble method Ensemble-S achieved a specificity score of 97.98% for clinically benign cases, 97.30% for the melanoma cases, and 98.58% for the seborrhoeic keratosis cases on ISIC-2017 testing set, exhibiting better performance than FrCN, FCNs, U-Net, and SegNet. |
| first_indexed | 2024-12-22T19:44:44Z |
| format | Article |
| id | doaj.art-384191e5c3544ccc9f099a54970eb8dc |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-22T19:44:44Z |
| publishDate | 2020-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-384191e5c3544ccc9f099a54970eb8dc2022-12-21T18:14:43ZengIEEEIEEE Access2169-35362020-01-0184171418110.1109/ACCESS.2019.29605048936444Skin Lesion Segmentation in Dermoscopic Images With Ensemble Deep Learning MethodsManu Goyal0https://orcid.org/0000-0002-9201-1385Amanda Oakley1https://orcid.org/0000-0002-9461-2790Priyanka Bansal2https://orcid.org/0000-0001-9108-7110Darren Dancey3https://orcid.org/0000-0001-7251-8958Moi Hoon Yap4https://orcid.org/0000-0001-7681-4287Department of Computing and Mathematics, Manchester Metropolitan University, Manchester, U.K.Waikato Clinical School, The University of Auckland, Hamilton, New ZealandDepartment of Computing and Mathematics, Manchester Metropolitan University, Manchester, U.K.Department of Computing and Mathematics, Manchester Metropolitan University, Manchester, U.K.Department of Computing and Mathematics, Manchester Metropolitan University, Manchester, U.K.Early detection of skin cancer, particularly melanoma, is crucial to enable advanced treatment. Due to the rapid growth in the number of skin cancers, there is a growing need of computerised analysis for skin lesions. The state-of-the-art public available datasets for skin lesions are often accompanied with a very limited amount of segmentation ground truth labeling. Also, the available segmentation datasets consist of noisy expert annotations reflecting the fact that precise annotations to represent the boundary of skin lesions are laborious and expensive. The lesion boundary segmentation is vital to locate the lesion accurately in dermoscopic images and lesion diagnosis of different skin lesion types. In this work, we propose the fully automated deep learning ensemble methods to achieve high sensitivity and high specificity in lesion boundary segmentation. We trained the ensemble methods based on Mask R-CNN and DeeplabV3+ methods on ISIC-2017 segmentation training set and evaluate the performance of the ensemble networks on ISIC-2017 testing set and PH2 dataset. Our results showed that the proposed ensemble methods segmented the skin lesions with Sensitivity of 89.93% and Specificity of 97.94% for the ISIC-2017 testing set. The proposed ensemble method Ensemble-A outperformed FrCN, FCNs, U-Net, and SegNet in Sensitivity by 4.4%, 8.8%, 22.7%, and 9.8% respectively. Furthermore, the proposed ensemble method Ensemble-S achieved a specificity score of 97.98% for clinically benign cases, 97.30% for the melanoma cases, and 98.58% for the seborrhoeic keratosis cases on ISIC-2017 testing set, exhibiting better performance than FrCN, FCNs, U-Net, and SegNet.https://ieeexplore.ieee.org/document/8936444/Skin cancerskin lesion segmentationensemble segmentation methodsdeep learningmelanomainstance segmentation |
| spellingShingle | Manu Goyal Amanda Oakley Priyanka Bansal Darren Dancey Moi Hoon Yap Skin Lesion Segmentation in Dermoscopic Images With Ensemble Deep Learning Methods IEEE Access Skin cancer skin lesion segmentation ensemble segmentation methods deep learning melanoma instance segmentation |
| title | Skin Lesion Segmentation in Dermoscopic Images With Ensemble Deep Learning Methods |
| title_full | Skin Lesion Segmentation in Dermoscopic Images With Ensemble Deep Learning Methods |
| title_fullStr | Skin Lesion Segmentation in Dermoscopic Images With Ensemble Deep Learning Methods |
| title_full_unstemmed | Skin Lesion Segmentation in Dermoscopic Images With Ensemble Deep Learning Methods |
| title_short | Skin Lesion Segmentation in Dermoscopic Images With Ensemble Deep Learning Methods |
| title_sort | skin lesion segmentation in dermoscopic images with ensemble deep learning methods |
| topic | Skin cancer skin lesion segmentation ensemble segmentation methods deep learning melanoma instance segmentation |
| url | https://ieeexplore.ieee.org/document/8936444/ |
| work_keys_str_mv | AT manugoyal skinlesionsegmentationindermoscopicimageswithensembledeeplearningmethods AT amandaoakley skinlesionsegmentationindermoscopicimageswithensembledeeplearningmethods AT priyankabansal skinlesionsegmentationindermoscopicimageswithensembledeeplearningmethods AT darrendancey skinlesionsegmentationindermoscopicimageswithensembledeeplearningmethods AT moihoonyap skinlesionsegmentationindermoscopicimageswithensembledeeplearningmethods |