Fine-Grained Classification of Unpigmented Skin Cancer from Paired Dermatoscopy Images
(2025) 23rd Scandinavian Conference on Image Analysis, SCIA 2025 In Lecture Notes in Computer Science 15726 LNCS. p.293-306- Abstract
Unpigmented skin cancer is the most prevalent form of cancer, and it burdens healthcare substantially even if it is not as aggressive as the more well-known malignant melanoma. Dermatoscopy images are commonly used for diagnosis, but differentiating between the many sub-diagnoses is a hard task. In this study we focus on these unpigmented cancers, performing both detection of basal cell carcinoma as well as fine-grained classification of its subclasses. We do this using a new dataset with more than 2’000 cases from a fair-skinned population. A deep learning model is specially designed for the task, handling pairs of polarised and non-polarised dermatoscopy images as input. We investigate transfer learning with different backbones as... (More)
Unpigmented skin cancer is the most prevalent form of cancer, and it burdens healthcare substantially even if it is not as aggressive as the more well-known malignant melanoma. Dermatoscopy images are commonly used for diagnosis, but differentiating between the many sub-diagnoses is a hard task. In this study we focus on these unpigmented cancers, performing both detection of basal cell carcinoma as well as fine-grained classification of its subclasses. We do this using a new dataset with more than 2’000 cases from a fair-skinned population. A deep learning model is specially designed for the task, handling pairs of polarised and non-polarised dermatoscopy images as input. We investigate transfer learning with different backbones as well as adding a mid-step of contrastive learning. The performance is compared to the accuracy of dermatologists on the subset of our test data where we have additional ground truth from histopathological diagnosis. This is the first study focusing on fine-grained classification of unpigmented lesions using only dermatoscopic images, and we reach a balanced accuracy of 39.4%, to be compared to 51.9% for dermatologists. This is a promising first step towards an algorithm to assist dermatologists in their work and we hope that this will open up for further studies on this interesting problem.
(Less)
- author
- Gummeson, Anna
LU
; Flood, Gabrielle
LU
; Nielsen, Kari LU
; Nätterdahl, Carolina LU
; Johansson, Fredrik LU
; Ingvar, Åsa LU
and Arvidsson, Ida LU
- organization
-
- Computer Vision and Machine Learning (research group)
- ELLIIT: the Linköping-Lund initiative on IT and mobile communication
- eSSENCE: The e-Science Collaboration
- LTH Profile Area: AI and Digitalization
- LUSCaR- Lund University Skin Cancer Research group (research group)
- Lund Melanoma Study Group (research group)
- LU Profile Area: Natural and Artificial Cognition
- LU Profile Area: Proactive Ageing
- Dermatology and Venereology (Lund)
- publishing date
- 2025
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- Deep learning, Dermatoscopy, Fine-grained classification, Unpigmented skin cancer
- host publication
- Image Analysis - 23rd Scandinavian Conference, SCIA 2025, Proceedings
- series title
- Lecture Notes in Computer Science
- editor
- Petersen, Jens and Dahl, Vedrana Andersen
- volume
- 15726 LNCS
- pages
- 293 - 306
- publisher
- Springer Science and Business Media B.V.
- conference name
- 23rd Scandinavian Conference on Image Analysis, SCIA 2025
- conference location
- Reykjavik, Iceland
- conference dates
- 2025-06-23 - 2025-06-25
- external identifiers
-
- scopus:105009823663
- ISSN
- 1611-3349
- 0302-9743
- ISBN
- 9783031959172
- DOI
- 10.1007/978-3-031-95918-9_21
- language
- English
- LU publication?
- yes
- additional info
- Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.
- id
- b642a2de-8ef8-4e55-8b8d-a661da1b0c88
- date added to LUP
- 2025-09-11 22:20:02
- date last changed
- 2025-09-26 03:27:54
@inproceedings{b642a2de-8ef8-4e55-8b8d-a661da1b0c88, abstract = {{<p>Unpigmented skin cancer is the most prevalent form of cancer, and it burdens healthcare substantially even if it is not as aggressive as the more well-known malignant melanoma. Dermatoscopy images are commonly used for diagnosis, but differentiating between the many sub-diagnoses is a hard task. In this study we focus on these unpigmented cancers, performing both detection of basal cell carcinoma as well as fine-grained classification of its subclasses. We do this using a new dataset with more than 2’000 cases from a fair-skinned population. A deep learning model is specially designed for the task, handling pairs of polarised and non-polarised dermatoscopy images as input. We investigate transfer learning with different backbones as well as adding a mid-step of contrastive learning. The performance is compared to the accuracy of dermatologists on the subset of our test data where we have additional ground truth from histopathological diagnosis. This is the first study focusing on fine-grained classification of unpigmented lesions using only dermatoscopic images, and we reach a balanced accuracy of 39.4%, to be compared to 51.9% for dermatologists. This is a promising first step towards an algorithm to assist dermatologists in their work and we hope that this will open up for further studies on this interesting problem.</p>}}, author = {{Gummeson, Anna and Flood, Gabrielle and Nielsen, Kari and Nätterdahl, Carolina and Johansson, Fredrik and Ingvar, Åsa and Arvidsson, Ida}}, booktitle = {{Image Analysis - 23rd Scandinavian Conference, SCIA 2025, Proceedings}}, editor = {{Petersen, Jens and Dahl, Vedrana Andersen}}, isbn = {{9783031959172}}, issn = {{1611-3349}}, keywords = {{Deep learning; Dermatoscopy; Fine-grained classification; Unpigmented skin cancer}}, language = {{eng}}, pages = {{293--306}}, publisher = {{Springer Science and Business Media B.V.}}, series = {{Lecture Notes in Computer Science}}, title = {{Fine-Grained Classification of Unpigmented Skin Cancer from Paired Dermatoscopy Images}}, url = {{http://dx.doi.org/10.1007/978-3-031-95918-9_21}}, doi = {{10.1007/978-3-031-95918-9_21}}, volume = {{15726 LNCS}}, year = {{2025}}, }