Convolutional neural networks in skin cancer detection using spatial and spectral domain
Abstract
Skin cancers are world wide deathly health problem, where significant life and cost savings could be achieved if
detection of cancer can be done in early phase. Hypespectral imaging is prominent tool for non-invasive screening.
In this study we compare how use of both spectral and spatial domain increase classification performance of
convolutional neural networks. We compare five different neural network architectures for real patient data. Our
models gain same or slightly better positive predictive value as clinicians. Towards more general and reliable
model more data is needed and collection of training data should be systematic.
Main Authors
Format
Conferences
Conference paper
Published
2019
Series
Subjects
Publication in research information system
Publisher
SPIE, The International Society for Optical Engineering
The permanent address of the publication
https://urn.fi/URN:NBN:fi:jyu-201904042072Use this for linking
Review status
Peer reviewed
ISSN
0277-786X
DOI
https://doi.org/10.1117/12.2509871
Conference
Photonics in Dermatology and Plastic Surgery
Language
English
Published in
SPIE conference proceedings
Is part of publication
Proceedings of SPIE Volume 10851 : Photonics in Dermatology and Plastic Surgery 2019
Citation
- Pölönen, I., Rahkonen, S., Annala, L., & Neittaanmäki, N. (2019). Convolutional neural networks in skin cancer detection using spatial and spectral domain. In B. Choi, & H. Zeng (Eds.), Proceedings of SPIE Volume 10851 : Photonics in Dermatology and Plastic Surgery 2019 (Article 108510B). SPIE, The International Society for Optical Engineering. SPIE conference proceedings, 10851. https://doi.org/10.1117/12.2509871
License
Copyright© Society of Photo-Optical Instrumentation Engineers (SPIE), 2019.