Objective: To evaluate the potential of a novel imaging technology, optical transfer diagnosis (OTD), for differentiation of benign from malignant pigmented melanocytic lesions.

Design: Patients with pigmented lesions suspicious for melanoma were referred for OTD. After scanning, lesions were biopsied for histopathologic examination, each by two separate dermatopathologists. To create morphologic-physiologic maps, the imaging system used the morphologic and physiologic parameters derived from prediction models of light absorption and scattering by chromophores such as hemoglobin, keratin, and melanin at different epidermal and dermal depths. The relative entropies were analyzed for output prediction of malignancy vs. nonmalignancy.

Setting: General dermatology clinic in a tertiary care academic medical center.

Patients: Fifty patients with suspected melanoma.

Intervention: OTD of pigmented lesions suspicious for melanoma, followed by biopsies for histopathologic examination.

Main Outcome Measures: Histopathologic confirmation of malignant lesions identified by OTD as melanoma.

Results: Sixty-three pigmented suspicious lesions were scanned before being biopsied for histopathologic examination by the two dermatopathologists. Of the 63 lesions, five were identified as melanoma and 58 were found to be benign (including three seborrheic keratoses and 55 melanocytic nevi). OTD was able to identify the malignant lesions with 100% sensitivity and 94.8-96.6% specificity.

Conclusions: Further study is indicated, but this technology is a promising adjunct to clinical skin cancer screening. Additionally, if the physiologic prediction models can be validated, OTD may facilitate the noninvasive study of some aspects of cutaneous physiology.

Download full-text PDF

Source
http://dx.doi.org/10.1111/j.1600-0846.2009.00367.xDOI Listing

Publication Analysis

Top Keywords

pigmented lesions
12
optical transfer
8
transfer diagnosis
8
lesions
8
lesions suspicious
8
suspicious melanoma
8
biopsied histopathologic
8
histopathologic examination
8
prediction models
8
malignant lesions
8

Similar Publications

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!