AI-Based Risk Classification and Histopathological Subtype Prediction of Basal Cell Carcinoma Using Dermoscopic Images (BCC-AI)

Risk Classification and Prediction of Histopathological Subtypes in Basal Cell Carcinoma Using a CNN-Based Artificial Intelligence Model on Dermoscopic Images

This retrospective observational study aims to develop and evaluate a convolutional neural network (CNN)-based artificial intelligence model for risk classification and histopathological subtype prediction of basal cell carcinoma (BCC) using clinical and dermoscopic images. Histopathologically confirmed BCC cases from a dermatology archive will be included. The primary objective is to assess the diagnostic performance of the CNN model in classifying BCC as low-risk or high-risk. Secondary objectives include predicting histopathological subtypes and comparing the model's performance with that of dermatology physicians. Histopathological diagnosis will serve as the reference standard. All archived data will be anonymized before analysis.

Study Overview

• Status: Recruiting
• Conditions: Basal Cell Carcinoma

Detailed Description

Basal cell carcinoma (BCC) is the most common skin malignancy and comprises histopathological subtypes with different biological behaviors, recurrence risks, and treatment implications. Accurate identification of high-risk and low-risk subtypes is important for clinical decision-making. Dermoscopy improves diagnostic accuracy in BCC; however, prediction of histopathological risk categories based solely on dermoscopic findings remains challenging.

This retrospective observational study will use archived clinical and dermoscopic images, histopathology reports, and clinical records of patients with histopathologically confirmed BCC. All data will be anonymized before analysis. Images containing identifiable patient information will be excluded.

A convolutional neural network (CNN)-based artificial intelligence model will be developed using clinical and dermoscopic images. Images will undergo preprocessing, including standardization of image size, normalization procedures, and removal of potentially identifiable information. The dataset will be divided into training, validation, and test sets while maintaining separation at the patient level to avoid data leakage.

The primary outcome is the diagnostic performance of the CNN model for classification of BCC into low-risk and high-risk histopathological groups. Secondary outcomes include prediction of histopathological subtypes and comparison of model performance with dermatologist assessments. Histopathological diagnosis will serve as the reference standard.

Model performance will be evaluated using accuracy, sensitivity, specificity, precision, recall, F1 score, and area under the receiver operating characteristic curve (ROC-AUC). Comparisons between the artificial intelligence model and physician assessments will be performed using appropriate statistical methods. Interobserver agreement may also be assessed when applicable.

• Study Type: Observational
• Enrollment (Estimated): 2500

Contacts and Locations

• Study Contact: Name: tugce nur izbudak kara, MD; Phone Number: +905395976598; Email: eizbudak@icloud.com

Study Locations

• Turkey (Türkiye)
  • Istanbul
    • Istanbul, Istanbul, Turkey (Türkiye), 34000
      • Recruiting
      • Istanbul Training and Research Hospital
      • Contact: tugce nur izbudak kara, MD; Phone Number: +905395976598; Email: eizbudak@icloud.com
      • Contact: Ayse Esra Koku Aksu, MD; Phone Number: +905059126069; Email: esraaksu@gmail.com
      • Principal Investigator: Ayse Esra Koku Aksu, MD
      • Sub-Investigator: Tugce Nur Izbudak Kara, MD
      • Sub-Investigator: Duygu Yamen, MD

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 patients with histopathologically confirmed basal cell carcinoma who have dermoscopic images of sufficient quality for artificial intelligence analysis and documented histopathological subtype information. Archived clinical and dermoscopic images collected at Istanbul Training and Research Hospital will be retrospectively analyzed.

Description

Inclusion Criteria:

• Patients with histopathologically confirmed basal cell carcinoma.
• Cases with a specified histopathological subtype.
• Availability of dermoscopic images with sufficient image quality and resolution for artificial intelligence analysis.

Exclusion Criteria:

• Cases without histopathological confirmation of basal cell carcinoma.
• Cases with unspecified histopathological subtype.
• Images with insufficient quality or resolution for artificial intelligence analysis.
• Cases without available dermoscopic images.

Study Plan

How is the study designed?

Number of groups / cohorts

2

Cohorts and Interventions

Group / Cohort
Low-Risk Basal Cell Carcinoma; Patients with histopathologically confirmed low-risk basal cell carcinoma, including nodular, superficial, pigmented, adenoid, solid, and nodulocystic subtypes. Clinical and dermoscopic images will be used for artificial intelligence-based risk classification and subtype prediction.
High-Risk Basal Cell Carcinoma; Patients with histopathologically confirmed high-risk basal cell carcinoma, including infiltrative, micronodular, morpheaform, and basosquamous subtypes. Clinical and dermoscopic images will be used for artificial intelligence-based risk classification and histopathological subtype prediction.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Accuracy of artificial intelligence-based classification of basal cell carcinoma risk groups	Diagnostic accuracy of the convolutional neural network model in distinguishing low-risk and high-risk basal cell carcinoma using dermoscopic images, compared with histopathological diagnosis as the reference standard.	Baseline

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Diagnostic accuracy (accuracy, sensitivity, specificity, F1-score and ROC-AUC) of convolutional neural network for histopathological subtype prediction of basal cell carcinoma using dermoscopic images	Diagnostic performance of the convolutional neural network in predicting histopathological subtypes of basal cell carcinoma from dermoscopic images compared with histopathological diagnosis (reference standard). Diagnostic accuracy will be assessed using accuracy, sensitivity, specificity, precision, F1-score and ROC-AUC.	baseline
Diagnostic accuracy (accuracy, sensitivity, specificity, F1-score and ROC-AUC) of artificial intelligence compared with dermatologists for basal cell carcinoma risk classification	Comparison of diagnostic performance between the artificial intelligence model and dermatologists in risk classification of basal cell carcinoma. Performance will be assessed using accuracy, sensitivity, specificity, precision, F1-score and ROC-AUC.	baseline

Collaborators and Investigators

• Sponsor: Istanbul Training and Research Hospital
• Investigators: Principal Investigator: Ayse Esra Koku Aksu, MD, Istanbul Training and Research Hospital

Study record dates

Study Major Dates

• Study Start (Actual): May 22, 2026
• Primary Completion (Estimated): May 22, 2027
• Study Completion (Estimated): May 22, 2027

Study Registration Dates

• First Submitted: June 19, 2026
• First Submitted That Met QC Criteria: June 29, 2026
• First Posted (Actual): June 30, 2026

Study Record Updates

• Last Update Posted (Actual): June 30, 2026
• Last Update Submitted That Met QC Criteria: June 29, 2026
• Last Verified: June 1, 2026

More Information

Terms related to this study

• Keywords: Skin Cancer; Artificial Intelligence; Basal Cell Carcinoma; Dermoscopy; Convolutional Neural Network; Histopathological Subtypes; Risk Classification
• Additional Relevant MeSH Terms: Neoplasms by Site; Neoplasms; Neoplasms by Histologic Type; Neoplasms, Glandular and Epithelial; Skin Diseases; Carcinoma; Neoplasms, Basal Cell; Skin and Connective Tissue Diseases; Skin Neoplasms; Carcinoma, Basal Cell
• Other Study ID Numbers: SBU-IEAH-DERM-BCCAI-163; IEAH-EC-163 (Other Identifier: Istanbul Training and Research Hospital Non-Interventional Clinical Research Ethics Committee)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: YES

Drug and device information, study documents

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