摘要
Background: In a pilot study, the three-point checklist of dermoscopy has been shown to represent a valid and reproducible tool with high sensitivity for the diagnosis of skin cancer in the hands of a small group of nonexperts. Objectives: To reevaluate these preliminary results in a large number of observers independently from their profession and expertise in dermoscopy. Methods: The study was conducted via the internet to provide worldwide access for participants. After a short webbased tutorial, the participants evaluated dermoscopic images of 165 (116 benign and 49 malignant) skin lesions (15 training and 150 test lesions). For each lesion participants scored the presence of the three-point checklist criteria (asymmetry, atypical network and blue-white structures). Kappa values, odds ratios, sensitivity, specificity and likelihood ratios were estimated. Results: Overall, 150 participants joined the study. The three-point checklist showed good interobserver reproducibility (kappa value: 0.53). Sensitivity for skin cancer (melanoma and basal cell carcinoma) was 91.0%and this value remained basically uninfluenced by the observers’professional profile. Only 20 participants lacking any experience in dermoscopy performed significantly more poorly, but the sensitivity was still remarkably high (86.7%) when considering that they were untrained novices in dermoscopy. The specificity was 71.9%and was significantly influenced by the profession, with dermatologists performing best. Conclusions: Our study confirms that the three-point checklist is a feasible, simple, accurate and reproducible skin cancer screening tool.
Background： In a pilot study, the three-point checklist of dermoscopy has been shown to represent a valid and reproducible tool with high sensitivity for the diagnosis of skin cancer in the hands of a small group of nonexperts. Objectives： To reevaluate these preliminary results in a large number of observers independently from their profession and expertise in dermoscopy. Methods： The study was conducted via the intemet to provide worldwide access for participants. After a short webbased tutorial, the participants evaluated dermoscopic images of 165 （116 benign and 49 malignant） skin lesions （15 training and 150 test lesions）. For each lesion participants scored the presence of the three-point checklist criteria （asymmetry, atypical network and blue-white .structures） . Kappa val- ues, odds ratios, sensitivity, specificity and likelihood ratios were estimated. Results： Overall, 150 participants joined the study. The three-point checklist showed good interobserver reproducibility （kappa value： 0. 53） . Sensitivity for skin cancer （melanoma and basal cell carcinoma） was 91.0% and this value remained basically uninfluenced by the observers＇ professional profile. Only 20 participants lacking any experience in dermoscopy performed significantly more poorly, but the sensitivity was still remarkably high （86.7%） when considering that they were untrained novices in dermoscopy. The specificity was 71.9% and was significantly influenced by the profession, with dermatologists performing best. Conclusions： Our study confirms that the three-point checklist is a feasible, simple, accurate and reproducible skin cancer screening tool.
