Quantitative Multispectral Imaging Differentiates Melanoma from Seborrheic Keratosis

Area of Science:

• Dermatology
• Medical Imaging
• Oncology

Background:

• Melanoma, a dangerous skin cancer, can be clinically mistaken for seborrheic keratosis (SK), a common benign skin lesion.
• Accurate differentiation is crucial for timely melanoma treatment and avoiding unnecessary procedures for benign SKs.

Purpose of the Study:

• To evaluate the efficacy of multispectral imaging in distinguishing melanoma from seborrheic keratosis.
• To identify quantitative imaging parameters that can aid in differentiating these two skin conditions.

Main Methods:

• Utilized a multispectral imaging device employing autofluorescence (405 nm) and diffuse reflectance (525 nm, 660 nm) imaging.
• Analyzed image intensity descriptors, including channel ratios, standard deviation, and min/max values.
• Assessed lesion patterns using particle analysis and developed an 'SK index'.

Main Results:

• Significantly higher intensity values were observed in seborrheic keratoses compared to melanoma, particularly in the autofluorescence channel.
• Seborrheic keratoses exhibited a higher number of particles with high fluorescence.
• The developed 'SK index' achieved a sensitivity of 91.9% and specificity of 57.0% in differentiating melanoma from SK.

Conclusions:

• Multispectral imaging analysis of quantitative parameters shows potential for distinguishing melanoma from seborrheic keratosis.
• This imaging technique could serve as a valuable screening tool for general physicians and non-experts in clinical practice.