Elastic scattering spectroscopy in the diagnosis of pigmented lesions: comparison with clinical and histopathological diagnosis

J. J. Scarisbrick; C. D. O. Pickard; Andrew C. Lee; Gavin M. Briggs; Kristie Johnson; Stephen G. Bown; Marco Novelli; M. R. S. Keshtgar; Irving J. Bigio; R. Yu

doi:10.1117/12.501181

8 October 2003 Elastic scattering spectroscopy in the diagnosis of pigmented lesions: comparison with clinical and histopathological diagnosis

J. J. Scarisbrick, C. D. O. Pickard, Andrew C. Lee, Gavin M. Briggs, Kristie Johnson, Stephen G. Bown, Marco Novelli, M. R. S. Keshtgar, Irving J. Bigio, R. Yu

Author Affiliations +

Proceedings Volume 5141, Diagnostic Optical Spectroscopy in Biomedicine II; (2003) https://doi.org/10.1117/12.501181
Event: European Conferences on Biomedical Optics 2003, 2003, Munich, Germany

Abstract

Successful management of malignant melanoma depends on early detection and diagnostic accuracy. However studies have found a diagnostic accuracy of malignant melanoma amongst dermatologists of only 80% compared with histological diagnosis, results are generally poorer with family practitioners and dermatology trainees. Elastic scattering spectroscopy (ESS) is a non-invasive procedure that utilises elastically scattered light. A related technique, reflectance spectrometry, may discriminate between benign and malignant pigmented lesions in vivo. Objectives: We have therefore assessed an ESS biopsy system in the diagnosis of melanocytic lesions in vivo and compared the results to both clinical and histopathological diagnosis. Patients/Methods: One hundred melanocytic lesions from 77 patients attending our pigmented lesion clinic between 1999 and 2001 or seen at our Melanoma Day were examined clinically and divided into benign, dysplastic or malignant lesions. ESS spectra were acquired at several points from each lesion and adjacent normal skin. Lesions were then excised and sent for histological examination. Results: Lesions were histologically classified as 12 malignant melanoma (3 in-situ), 14 dysplastic naevi and 57 benign naevi (a further 17 clinically benign naevi were included which were not selected for excision). Clinical examination had a sensitivity of 75% and specificity of 91% for detection of malignant melanoma from other melanocytic lesions and overall diagnostic accuracy of differentiating melanocytic lesions of 71%. ESS spectral diagnosis classified by linear discriminant analysis indicated a sensitivity of 84% and specificity of 65% for detection of malignant melanoma from other melanocytic lesions. However the ESS spectral diagnosis included several readings per lesion and clinically this is not relevant. Therefore an ESS per lesional diagnosis was taken as the most aggressive diagnosis. This had a sensitivity of 100% and specificity of 75% for detection of malignant melanoma from other melanocytic lesions and overall diagnostic accuracy of differentiating melanocytic lesions of 69%. Discussion: These data suggest that ESS may be capable of differentiating malignant melanoma, dysplastic and benign naevi with a high sensitivity and provide adjuvant information to clinical examination by a dermatologist, which could be useful in the triage of melanocytic lesions.

Citation Download Citation

J. J. Scarisbrick, C. D. O. Pickard, Andrew C. Lee, Gavin M. Briggs, Kristie Johnson, Stephen G. Bown, Marco Novelli, M. R. S. Keshtgar, Irving J. Bigio, and R. Yu "Elastic scattering spectroscopy in the diagnosis of pigmented lesions: comparison with clinical and histopathological diagnosis", Proc. SPIE 5141, Diagnostic Optical Spectroscopy in Biomedicine II, (8 October 2003); https://doi.org/10.1117/12.501181

ACCESS THE FULL ARTICLE