A handheld Raman spectroscopy system with machine learning distinguished normal skin, basal cell carcinoma (BCC), and squamous cell carcinoma (SCC) with 84% accuracy.
2
Biopsy and microscopic examination remain the standard for diagnosing BCC and SCC, which can resemble benign lesions.
3
The study utilized a mobile system with a 785-nm laser, examining over 50 tissue samples and producing nearly 1,000 Raman spectra.
4
K-nearest neighbors and support vector machine models achieved the highest accuracy, with sensitivity of 78.7% and specificity of 88.6%.
5
The study focused on analytical performance rather than clinical diagnostic accuracy, with results below the desired 90% threshold for clinical use.
