Diagnosis of melanoma involves physical examination of the lesion with many dermatologists adjunctively employing dermoscopes. The rate of misdiagnosis of melanoma remains significant, along with a high rate of referral to biopsy. Elucid Labs (Waterloo, Ontario) has developed a novel handheld, digital dermoscope with accompanying visualization and analysis software - the Artificial Intelligence Dermatology Assistant (AIDA™). Apart from collecting conventional demoscopic images, AIDA also collects images at various spectral bands. The aim of this study is to understand and quantify the value of this novel adjunctive information for dermatologists diagnosing atypical skin lesions.
Patients presenting with atypical skin lesions will undergo assessment by an investigator as per their standard clinical practice (not utilizing AIDA™). If a lesion meeting the inclusion-exclusion criteria is referred for biopsy, informed consent will be obtained and the subject will be enrolled. Subjects will then have images acquired by the AIDA™ system. All lesions scheduled for biopsy (Subgroup A) will be imaged along with at most 2 additional lesions meeting inclusion/exclusion criteria but not referred for biopsy (Subgroup B). For each lesion imaged using AIDA™, the investigator will manually segment the lesion image and list any lesion features which contributed to their recommendation to biopsy or not biopsy. The investigator will first score the lesion according to the ABCD rule using the standard dermoscopy image displayed. They will then state their diagnosis (malignant, dyplastic, or benign) and their diagnostic confidence using a visual analog scale. Once standard demoscopy diagnosis has been collected, the process will be repeated with the use of AIDA™ software outputs. Investigators will also provide an estimate of lesion depth based on AIDA™ depth images. All biopsy results will be recorded by the pathologist. Histopathology determination will be used as the definitive diagnosis of either positive (malignant/dysplastic) or negative (benign). Complete de-identified pathology reports may also be collected.
Study Type
INTERVENTIONAL
Allocation
NA
Purpose
DIAGNOSTIC
Masking
NONE
Enrollment
4,000
The Artificial Intelligence Dermatology Assistant (AIDA™) collects conventional demoscopic images and images at various spectral bands. Following image acquisition, the AIDA™ software presents users with (1) similar lesion images from the International Skin Imaging Collaboration archive, (2) Hypodermoscopy™ images, and (3) images providing an indication of lesion depth, based on the spectral data.
Total Skincare Centre
Calgary, Alberta, Canada
Sensitivity of in-clinic dermatologist diagnosis using AIDA™ compared to standard dermoscopy and physical examination alone
The investigator will review the standard dermoscopy image for the lesion, score it according to the ABCD rule, and state their diagnosis (malignant, dysplastic, or benign) and their diagnostic confidence using a visual analog scale. Subsequently, the investigator will review the AIDA™ outputs and again state their diagnosis and confidence. The sensitivity of those in-clinic diagnoses will be determined by using the definitive diagnoses established in the histopathology reports. The sensitivity of a diagnostic technique determines the probability of a positive test result in a person who has the disease. This is defined according to the equation: TP/(TP + FN) . In this equation, TP is the number of true-positive and FN is the number of false-negative results.
Time frame: Average of 4 weeks
Specificity of in-clinic dermatologist diagnosis using AIDA™ compared to standard dermoscopy and physical examination alone
The investigator will review the standard dermoscopy image for the lesion, score it according to the ABCD rule, and state their diagnosis (malignant, dysplastic, or benign) and their diagnostic confidence using a visual analog scale. Subsequently, the investigator will review the AIDA™ outputs and again state their diagnosis and confidence. The specificity of those in-clinic diagnoses will be determined by using the definitive diagnoses established in the histopathology reports. The specificity of a diagnostic technique refers to the probability of a negative test result in a person who does not have the disease according to the equation: TN/(TN + FP). In this equation, TN is the number of true-negative and FP is the number of false-positive results.
Time frame: Average of 4 weeks
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.