Actively Recruiting
Artificial Intelligence Analysis of Basal Cell Carcinoma Risk and Subtypes Using Dermoscopic Images
Led by Istanbul Training and Research Hospital · Updated on 2026-06-30
2500
Participants Needed
1
Research Sites
N/A
Total Duration
AI-Summary
What This Trial Is About
Researchers are evaluating an artificial intelligence model based on convolutional neural networks CNN to classify risk and predict histopathological subtypes of basal cell carcinoma BCC, the most common skin cancer. The study uses archived clinical and dermoscopic images from patients with histopathologically confirmed BCC to improve clinical decision-making by distinguishing between low-risk and high-risk subtypes. This observational research aims to compare the AI models diagnostic performance with that of dermatologists using histopathology as the reference standard. The study involves analyzing anonymized archived images and clinical data after preprocessing steps like image size standardization and normalization. The dataset will be divided into training, validation, and test groups to develop and evaluate the CNN model. Two groups of BCC cases are included low-risk subtypes such as nodular and superficial, and high-risk subtypes such as infiltrative and morpheaform. The models ability to classify risk groups and predict subtypes will be assessed. Participants involvement includes using existing clinical and dermoscopic images without new interventions. The AI models accuracy, sensitivity, specificity, precision, recall, F1 score, and ROC-AUC will be measured. Comparisons with dermatologist assessments and interobserver agreement analyses will also be performed. All data are anonymized to protect patient privacy, and no additional clinical visits or treatments are involved.
CONDITIONS
Brief Title
AI-Based Risk Classification and Histopathological Subtype Prediction of Basal Cell Carcinoma Using Dermoscopic Images
Research Team
T
Tugce Nur Izbudak Kara, MD
Not the Right Trial for You?
Explore thousands of other clinical trials that might be a better match.
Sign up to get personalized trial recommendations delivered to your inbox.
Already have an account? Log in here