BACKGROUND: There are potential clinical benefits if non-invasive methods can be used to diagnose or exclude melanoma. OBJECTIVES: We investigated high-resolution ultrasound (HRU) as a potential non-invasive diagnostic aid for pigmented skin lesions. METHODS: Using a 20-MHz ultrasound B-scan imaging system interfaced to a computer, we assessed acoustic shadowing and entry echo line enhancement (EEE) for 29 basal cell papillomas (BCPs) and 25 melanomas. Acoustic shadowing was estimated by the dermal echogenicity ratio (DER), comparing mean echogenicity below the lesion with that of adjacent dermis. Histological features were scored independently. RESULTS: DER < 3 correctly distinguished melanoma from BCP with 100% sensitivity and 79% specificity. Specificity increased to 93% if the presence of EEE was included as a discriminator. Shadowing correlated most significantly with histological extent of hyperkeratosis (P < 0.0001). Consequently, this method falsely identified non-keratotic acanthotic BCP (n = 3) as melanoma. Highly significant differences between benign naevi (n = 15) and melanomas (n = 24) were found. The SD of retrolesional echogenicity was higher for naevi than melanomas (P < 0.0001), but such an analysis was poorly specific for the diagnosis of melanoma (30%). CONCLUSIONS: Overall, HRU has considerable potential as a high-performance screening tool to assist in the discrimination between BCP, but not benign naevi, and melanoma. In particular, it may be possible to exclude melanoma with 100% certainty in the differentiation of BCP from melanoma.
BACKGROUND: There are potential clinical benefits if non-invasive methods can be used to diagnose or exclude melanoma. OBJECTIVES: We investigated high-resolution ultrasound (HRU) as a potential non-invasive diagnostic aid for pigmented skin lesions. METHODS: Using a 20-MHz ultrasound B-scan imaging system interfaced to a computer, we assessed acoustic shadowing and entry echo line enhancement (EEE) for 29 basal cell papillomas (BCPs) and 25 melanomas. Acoustic shadowing was estimated by the dermal echogenicity ratio (DER), comparing mean echogenicity below the lesion with that of adjacent dermis. Histological features were scored independently. RESULTS: DER < 3 correctly distinguished melanoma from BCP with 100% sensitivity and 79% specificity. Specificity increased to 93% if the presence of EEE was included as a discriminator. Shadowing correlated most significantly with histological extent of hyperkeratosis (P < 0.0001). Consequently, this method falsely identified non-keratotic acanthotic BCP (n = 3) as melanoma. Highly significant differences between benign naevi (n = 15) and melanomas (n = 24) were found. The SD of retrolesional echogenicity was higher for naevi than melanomas (P < 0.0001), but such an analysis was poorly specific for the diagnosis of melanoma (30%). CONCLUSIONS: Overall, HRU has considerable potential as a high-performance screening tool to assist in the discrimination between BCP, but not benign naevi, and melanoma. In particular, it may be possible to exclude melanoma with 100% certainty in the differentiation of BCP from melanoma.
Authors: Jeffrey A Ketterling; Orlando Aristizábal; Daniel H Turnbull Journal: IEEE Trans Ultrason Ferroelectr Freq Control Date: 2006-07 Impact factor: 2.725
Authors: Jeffrey A Ketterling; Orlando Aristizábal; Daniel H Turnbull; Frederic L Lizzi Journal: IEEE Trans Ultrason Ferroelectr Freq Control Date: 2005-04 Impact factor: 2.725