CNN-Based AI Versus Physicians for Solitary Skin Lesion Diagnosis

Comparison of a CNN-Based Artificial Intelligence Model With Dermatologists and Non-Dermatologist Physicians in the Diagnosis of Solitary Skin Lesions

The goal of this observational study is to evaluate the diagnostic accuracy of a CNN-based artificial intelligence model in patients with solitary skin lesions. The main questions it aims to answer are:

• What is the diagnostic performance (sensitivity and specificity) of the CNN-based model in identifying solitary skin lesions using macroscopic clinical images?
• How does the diagnostic accuracy of the CNN-based model compare with the evaluations performed by dermatologists and non-dermatologist physicians?

Researchers will compare the AI model's diagnostic outputs to the independent evaluations of dermatologists and non-dermatologist physicians to see if the AI model can achieve a diagnostic performance comparable to or better than human clinicians.

Participants (physicians acting as clinical readers) will:

• Independently review a predefined set of anonymized macroscopic clinical images sourced from a retrospective patient archive.
• Provide a primary diagnosis for each lesion based solely on the images, without access to patient history or histopathological results.
• Submit their assessments to be compared against the gold standard (histopathological diagnosis) and the AI model's results.

Study Overview

• Status: Active, not recruiting
• Conditions: Skin Neoplasms; Skin Neoplasm Malignant; Solitary Skin Lesions

Detailed Description

This study is a retrospective, observational diagnostic accuracy study designed to evaluate the performance of a convolutional neural network (CNN)-based artificial intelligence model in the assessment of solitary skin lesions using macroscopic clinical images.

Macroscopic clinical images of solitary skin lesions with histopathological or clinically confirmed diagnoses will be retrospectively retrieved from the dermatology image archive of Istanbul Training and Research Hospital. All images and associated clinical documents will be anonymized prior to analysis, and any identifying visual or textual information will be removed. Data processing and analysis will be conducted in a secure, institution-based environment with restricted access limited to the study team.

A CNN-based artificial intelligence model will be developed using supervised learning techniques. Image preprocessing steps will include resizing to standardized input dimensions, color normalization, and removal of regions containing potentially identifiable information. The dataset will be partitioned into training, validation, and test subsets to enable model development, hyperparameter optimization, and independent performance evaluation. Model training and evaluation will be implemented using the PyTorch deep learning framework.

The diagnostic performance of the CNN-based model will be evaluated using standard classification metrics and will be compared with the independent assessments of dermatologists, dermatology residents, and non-dermatologist physicians who evaluate the same set of anonymized images without access to additional clinical or histopathological information. Comparative analyses will be performed to assess differences in diagnostic performance and agreement between the artificial intelligence model and physician groups.

• Study Type: Observational
• Enrollment (Estimated): 17625

Contacts and Locations

Study Locations

• Turkey (Türkiye)
  • Fatih
    • Istanbul, Fatih, Turkey (Türkiye), 34098
      • S.B.Ü. İstanbul Eğitim ve Araştırma Hastanesi

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child; Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

The study population consists of licensed physicians, including dermatologists, dermatology residents, and non-dermatologist physicians, who independently evaluate anonymized macroscopic clinical images of solitary skin lesions for diagnostic assessment.

Description

Inclusion Criteria:

• Patients who have provided informed consent for the use of their clinical images in scientific research.
• Clinical images with a resolution exceeding 224x224 pixels, ensuring compatibility with the artificial intelligence architecture.
• Retrospective records of solitary skin lesions with confirmed diagnoses.

Exclusion Criteria:

• Patients who have not consented to the use of their clinical photographs for research purposes.
• Images containing potentially identifiable personal information or visual features that compromise patient anonymity.
• Images with a resolution lower than 224x224 pixels or poor diagnostic quality (e.g., blurring, significant occlusion).
• Duplicate images or entries for the same lesion.

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort

Development and Validation Cohort; "This is a single-arm retrospective study consisting of 17,625 archived clinical records with confirmed histopathological diagnoses. The cohort will serve as the primary dataset for AI model development. A specific subset of the test dataset will be independently evaluated by a panel of dermatologists and non-dermatologist physicians through a multiple-choice diagnostic task. The AI model's performance will be compared against both the gold-standard histopathological results and the diagnostic accuracy of the human observers."

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Diagnostic accuracy of the CNN-based artificial intelligence model	The diagnostic accuracy of the convolutional neural network (CNN)-based artificial intelligence model in the diagnosis of solitary skin lesions will be evaluated using accuracy and area under the receiver operating characteristic curve (ROC-AUC) values based on macroscopic clinical images.	Baseline (Retrospective data analysis will be completed within 4 months)

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Difference in diagnostic performance between the CNN-based model and dermatologists	The difference in diagnostic performance between the CNN-based artificial intelligence model and dermatologists will be evaluated based on accuracy metrics using the same set of macroscopic clinical images.	Baseline (Expected completion within 5 months)
Difference in diagnostic performance between the CNN-based model and non-dermatologist physicians	The difference in diagnostic performance between the CNN-based artificial intelligence model and non-dermatologist physicians will be evaluated based on accuracy metrics using the same image set.	Baseline (Expected completion within 5 months)
Sensitivity, specificity of the CNN-based model and physician groups	Sensitivity, specificity of the CNN-based artificial intelligence model and physician groups will be calculated and compared in the evaluation of solitary skin lesions.	Baseline (Expected completion within 5 months)
F1-score of the CNN-based model and physician groups	F1-score values of the CNN-based artificial intelligence model and physician groups will be calculated and compared in the evaluation of solitary skin lesions.	Baseline (Expected completion within 5 months)

Collaborators and Investigators

• Sponsor: Istanbul Training and Research Hospital
• Investigators: Study Chair: Ayşe Esra Koku Aksu, MD, Sağlık Bilimleri Üniversitesi İstanbul Eğitim ve Araştırma Hastanesi

Study record dates

Study Major Dates

• Study Start (Actual): January 15, 2026
• Primary Completion (Estimated): March 30, 2026
• Study Completion (Estimated): May 31, 2026

Study Registration Dates

• First Submitted: February 10, 2026
• First Submitted That Met QC Criteria: February 16, 2026
• First Posted (Actual): February 17, 2026

Study Record Updates

• Last Update Posted (Actual): February 19, 2026
• Last Update Submitted That Met QC Criteria: February 17, 2026
• Last Verified: February 1, 2026

More Information

Terms related to this study

• Keywords: Artificial Intelligence; Diagnostic Accuracy; Dermatology; Deep Learning; Convolutional Neural Network; Computer-Aided Diagnosis; Solitary Skin Lesions; Clinical Images; Physician Comparison; Dermatologists; Non-Dermatologist Physicians; Physician vs AI Performance
• Additional Relevant MeSH Terms: Neoplasms by Site; Neoplasms; Skin Diseases; Skin and Connective Tissue Diseases; Skin Neoplasms
• Other Study ID Numbers: SBU-IEAH-KAEK-20260109-19

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO
• IPD Plan Description: Individual participant data will not be shared due to institutional data protection policies and the use of retrospectively collected, anonymized clinical images. Data are stored in a secure institutional environment and are accessible only to the study team.

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No