Photo Credit: iStock.com/kali9
The following is a summary of “DERM-SUCCESS FDA Pivotal Study: A Multi-Reader Multi-Case Evaluation of Primary Care Physicians’ Skin Cancer Detection Using AI-Enabled Elastic Scattering Spectroscopy,” published in the May 2025 issue of Journal of Primary Care & Community Health by Ferris et al.
Elastic Scattering Spectroscopy (ESS), combined with an artificial intelligence–enabled handheld device, was developed to help primary care physicians (PCPs) distinguish benign from malignant skin lesions without surgical biopsy, aiding early oncology detection.
Researchers conducted a retrospective study to compare the diagnostic and management performance of PCPs with and without the use of the ESS device in detecting skin cancer.
They carried out a clinical utility study in which 108 PCPs assessed 100 skin lesion cases, including 50 with device output and 50 without. For each case, PCPs gave a diagnosis, management decision, and confidence level first without and then with the ESS device output. Sensitivity, specificity, area under the curve (AUC), and confidence were compared before and after using the device.
The results showed that with device-assisted visual assessment, diagnostic sensitivity improved significantly from 71.1% to 81.7% (P = .0085), and referral sensitivity increased from 82.0% to 91.4% (P = .0027) compared to visual assessment alone. Diagnostic specificity declined from 60.9% to 54.7% (P = .1896), while referral specificity decreased from 44.2% to 32.4% (P = .0256). Overall management performance, measured by area under the curve (AUC), rose from 0.708 to 0.762, and for cases with low physician confidence, AUC increased from 0.567 to 0.682. The proportion of physicians reporting high confidence in their management decisions grew from 36.8% to 53.4%.
Investigators concluded that the use of the ESS device significantly enhanced PCPs’ diagnostic and management sensitivities, overall performance, and confidence in evaluating and managing skin lesions.
Create Post
Twitter/X Preview
Logout