Code used for the robust statistical analysis of Hyperspectral signatures obtained from Non-Melanoma Skin Cancer patients
Author
Lloyd A. Courtenay
ORCID
https://orcid.org/0000-0002-4810-2001
Current Afiliations:
Universidad de Salamanca [USAL]
This code has been designed for the open-source free R programming languages.
This code was designed and prepared for the study by: Courtenay, L.A.; González-Aguilera, D.; Lagüela, S.; Del Pozo, S.; Ruiz-Mendez, C.; Barbero-García, I.; Román-Curto, C.; Cañueto, J.; Santos-Durán, C.; Cardeñozo-Álvarez, M.E.; Roncero-Riesco, M.; Hernandez-Lopez, D.; Guerrero-Sevilla, D.; Rodríguez-Gonzalvez, P. (2021) Hyperspectral Imaging and Robust Statistics in Non-Melanoma Skin Cancer Analysis, Biomedical Optics Express, 12(8):5107-5127, doi: 10.1364/BOE.428143
All of the research included within the present repository was collected as part of the HYPER-SKINCARE research project, funded by the Junta de Castilla y Leon with project reference number: GRS 1837/A/18.
TIDOP research group website: http://tidop.usal.es/
IBSAL research group website: https://ibsal.es/en/
The present repository contains:
- Dataset of Hyperspectral Signatures
- Comma delimited table containing all the hyperspectral signatures obtained from cutaneous Squamous Cell Carcinoma (SCC) patients, as well as Basal Cell Carcinoma (BCC) patients.
- Patient ID's have been included so as to differentiate between signatures obtained from the same patients, nevertheless, so as to ensure patient animosity, no further information has or will be provided.
- Sensor Band Information
- A list of each of the Headwall Nano-Hyperspec's bands, their frequencies, and corresponding colours of visible light where applicable.
- Statistical Results
- A folder containing all of the statistical results and values calculated for the present study, and consequently used to create the associated manuscipt/article.
- R Code for Robust Statistical Analysis
- All the R code used to process the corresponding hyperspectral signatures and extract the statistical results
- All of the code used to produce graphs and figures included within the study
The present repository also contains some unpublished data closely linked to a follow up article employing computational learning for the processing of these signatures. This complementary unpublished data was calculated using univariate logistic regression algorithms to calculate the receiver operating characteristic for each wavelength seperately. The corresponding Area Under Curves have been included within a table. This was performed for additional feature selection, and has also been incorporated into the JavaScript code to produce the "Optimal Window" figure.
Please cite this repository as:
Courtenay (2021) Code and Data for the HYPER-SKINCARE project and paper titled 'Hyperspectral Imaging and Robust Statistics in Non-Melanoma Skin Cancer Analysis'. https://github.com/LACourtenay/HyperSkinCare_Statistics
Comments, questions, doubts, suggestions and corrections can all be directed to L. A. Courtenay at the email provided above.