Nevus Doctor Clinical Decision Support

N/ACompletedNCT02012751

University Hospital of North Norway49 enrolled

Overview

The "Nevus doctor" is a dermatoscopy based computer decision support tool to assist general practitioners (GPs) in the classification of pigmented skin lesions (PSLs). The aim of the program "Nevus doctor" is to help GPs increase their diagnostic accuracy, in particular regarding the selection of suspicious PSLs that need biopsy or referral to specialist health care for further assessment. The aim of this study is to investigate the diagnostic performance of the decision support tool in a primary care setting in Norway. We hypothesize that the diagnostic accuracy of the computer program "Nevus doctor" is better than the performance of the GPs.

In the trial GPs examine patients who attend a primary health care center. A given PSL is first assessed clinically by the GP and afterwards photographed. A clinical photograph as well as a dermatoscopic photograph is taken, the latter using a dermatoscope attached to the lens of the camera. The dermatoscopic image is then processed by the computer program. The performance of "Nevus doctor" is compared with the GPs' assessment. A dermatologist reviews all cases based on the clinical and dermatoscopic photographs and this assessment serves as the reference standard.

Study Type

INTERVENTIONAL

Allocation

Purpose

DIAGNOSTIC

Masking

NONE

Enrollment

Conditions

Nevus Melanoma

Outcomes

Primary Outcomes

Number of correct test results

The computer program tests whether the appearance of a skin lesion is suggestive of skin cancer. There are two possible test outcomes: "suspicious" or "not suspicious". The reference standard is the assessment done by a dermatologist. The number (and percentage) of skin lesions correctly classified as "suspicious" or "not suspicious" is reported. Furthermore sensitivity, specificity, positive and negative predictive values are calculated.

Time frame: Up to 2 weeks from enrollment

Secondary Outcomes

Patterns regarding false test results

All cases with false test results are described in order to identify any pattern that can help to explain the false test result: correlation with specific types of skin lesions, poor image quality, gender and age of the patients.