Background Several publications suggest that the increasing melanoma incidence may partly be caused by histopathological overdiagnosis. Pathologists provide the current gold standard in skin lesion diagnostics, most often primarily based on the interpretation of histological slides. Still, it has been suggested that pathologists' diagnostic accuracy and confidence could be improved if they gained access to additional clinical information and in-vivo clinical and dermoscopic images of melanocytic tumors. However, it can be challenging to interpret clinical and dermoscopic images, and mastery typically requires more than six years of clinical experience. This learning journey can be significantly shortened if the trainee receives comprehensive training in pattern recognition for dermoscopy and clinical images, including immediate, accurate, and individualized feedback and access to a library with a large selection of skin lesion cases. This study examines the effect of digital training for pathologists in interpreting dermoscopic and clinical skin tumor images.
Former studies have only focused on melanocytic lesions. Still, most pathologists will receive both melanocytic and pigmented non-melanocytic lesions (seborrheic keratoses, dermatofibromas, etc.) due to clinical suspicion of melanoma. This study includes an un-filtered selection of 211 clinically melanoma-suspect skin lesions excised at a specialized surgical department; this material is named the DAHT cases.
Aim:
To examine how case-based online training in interpreting clinical and dermoscopic images affects a pathologist's ability to diagnose skin tumors.
Method Case Database
Lesion data were collected from patients between 02.11.2020 and 22.01.2021 at the Department of Plastic Surgery, Herlev Hospital. Requirements for eligibility for the current study are:
The patient was referred through the clinical cancer pathways for melanoma. The lesion was excised upon evaluation by the plastic surgeon.
Patients received oral and written information about the project and were asked to sign a consent form before participation. The participation did not affect the included patients' treatment, diagnostics, or follow-up. Upon consent, the following information was collected for each lesion:
Clinical image Dermoscopic image Patients´ CPR-number (personal ID-number) Sex and age of the patient Location of skin tumor (on a 3D avatar) Medical history (former treatment, congenital nevi, if pregnant, time of appearance of skin lesion, change in appearance, symptoms, former melanoma or other skin diseases, family history of melanoma, sun exposure within the last six months)
After excision, the specimen was prepared for pathological examination and a representative hematoxylin-eosin stain and, if available, a MelanA stain for each skin lesion was chosen for the study by an experienced dermatopathologist. These stains were subsequently digitized and coupled with the relevant information (dermoscopic and clinical image, tumor location, sex, age, lesion information, etc.). The CPR number was deleted, rendering the case anonymous. Each case is stored in a database under a random anonymous ID number.
Web-based IT platform
To maximize the number of pathologists that can be included in the study, the investigators have developed an IT platform for the trial (The DAHT platform). The platform enables the following features:
Sign-in Automated randomization Login Case presentation Diagnosis of cases and subquestions Tracking
The diagnostic options will be based on the standardized MPATH-Dx version 2.0 with additional non-melanocytic options based on the most common diagnoses to be excised due to suspicion of malignancy. After diagnosing the lesion, participating pathologists will rate their confidence and difficulty in the chosen diagnosis on a 6-step Likert scale similar to the one used in the MPATH-Dx system. They will also be asked whether they want a second opinion, if they need additional stains or tests, and which tests/stains. All cases will be presented in a randomized order unique to each participant.
The tracking feature will enable search pattern analysis, including time per diagnosis and percentage of time spent looking at the histology stain, dermoscopic image, clinical image, and clinical information.
Executive phase
General pathologists and dermatopathologists from Denmark, Australia, and other countries will be enrolled between 1.09.2023 and 1.12.2023. All participants will be sent an email informing them about the trial and the handling of their data before signing a digital consent. Upon inclusion, each enrolled pathologist will be asked to fill out a digital sign-in form with the following variables:
Name E-mail address (including verification) Age Sex Country Subspecialization (general pathologist or dermatopathologist) Number of years interpreting skin lesions Caseload from melanocytic lesions per month Percentage of caseload from melanocytic lesions Number of second opinion assessments per month Number of second opinion referrals per month Former training in dermoscopy (yes/no) Perceived relevance of dermoscopic images during histopathological evaluation on a 5-step Likert scale Routine with the use of digitized slides
Following the sign-in, all participants will be asked to answer a dermoscopy test consisting of 25 formerly validated test items. They will automatically be randomized (allocation ratio 1:1) to either the intervention or control group. Participants in the intervention group will receive immediate access to a previously developed digital (tablet or smartphone) educational platform for the dermoscopy of melanocytic lesions, watch educational videos on the correlation between dermoscopy and histology, and then answer the dermoscopy test again before diagnosing the lesions in this study. Participants in the control group will not be given access to the learning intervention but will diagnose the lesions from this study immediately after signing up and taking the dermoscopy test. Each participant will be asked to assess a minimum of 50 cases within 21 days, preferably all 211 cases.
Statistics:
The accuracy will be measured as the number of correct answers (validated by a former DAHT study: AISC-DAHT, consensus agreement) divided by the total number of answers. Inter-rater reliability will be measured using generalizability theory that allows calculations in nested and unbalanced designs. Mean time spent per diagnosis, perceived difficulty, and diagnostic confidence during the evaluation will be compared between the study groups using independent t-tests. A post hoc ANOVA will compare the time spent per diagnosis and the confidence of pathologists grouped by time spent training in dermoscopy.
