Incorporating demographics into a skin cancer diagnosis algorithm for Raman spectroscopy improves diagnostic specificity (Conference Presentation)

Jianhua Zhao; Haishan Zeng; Sunil Kalia; Harvey Lui

doi:10.1117/12.2513779

4 March 2019 Incorporating demographics into a skin cancer diagnosis algorithm for Raman spectroscopy improves diagnostic specificity (Conference Presentation)

Jianhua Zhao, Haishan Zeng, Sunil Kalia, Harvey Lui

Author Affiliations +

Proceedings Volume 10851, Photonics in Dermatology and Plastic Surgery 2019; 1085102 (2019) https://doi.org/10.1117/12.2513779
Event: SPIE BiOS, 2019, San Francisco, California, United States

Abstract

Background & objective: Skin cancer is a very common malignancy that occurs more frequently in fair skin and older individuals. Raman spectroscopy is a non-invasive optical technique that can be used as an adjunct for skin cancer diagnosis. The objective of this study is to evaluate whether incorporating patient demographics can improve skin cancer diagnosis based on Raman spectroscopy. Patients & Methods: Raman spectra of 731 lesions and their respective adjacent normal skin were measured in vivo using a real-time Raman spectrometer. The lesions were divided into skin cancers (including malignant melanoma, basal cell carcinoma, squamous cell carcinoma and precancerous lesion - actinic keratosis, n = 340) and benign skin lesions (including pigmented nevi and seborrheic keratosis, n = 391). Patient age, gender, skin type and location of the lesion were incorporated into the analysis. Multivariate statistical analysis including principal component and general discriminant analysis (PC-GDA) and partial least squares (PLS) are used for skin cancer discrimination based on leave-one-out cross-validation. Results: The posterior probability of being a cancer is significantly dependent on gender, age and location of the lesion (p<0.05) but independent of skin type (p>0.05). The area under the receiver operating characteristic curve (ROC) is increased from 0.905 (95%CI: 0.884-0.927) to 0.932 (95%CI: 0.919-0.945) after taking into account demographics. Correspondingly, the specificity is increased from 43.2% to 50.1% at sensitivity of 99%; and from 73.4% to 77.5% at sensitivity of 90%. Conclusions: The specificity is increased after incorporating demographics into the algorithm for skin cancer diagnosis based on Raman spectroscopy.

Conference Presentation

Citation Download Citation

Jianhua Zhao, Haishan Zeng, Sunil Kalia, and Harvey Lui "Incorporating demographics into a skin cancer diagnosis algorithm for Raman spectroscopy improves diagnostic specificity (Conference Presentation)", Proc. SPIE 10851, Photonics in Dermatology and Plastic Surgery 2019, 1085102 (4 March 2019); https://doi.org/10.1117/12.2513779

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available