None listed
Conditions
Interventions
None listed
Sponsors
Samsung Medical Center
Eligibility
Sex/Gender
All
Inclusion criteria
Inclusion criteria: Inclusion Criteria (Applicable to All Three Participating Centers) 1. Patients who undergo upper gastrointestinal endoscopy for clinical care at participating institutions during the study period and for whom endoscopic image data are naturally generated during the examination 2. Adult patients (aged 19 years or older) who undergo endoscopic submucosal dissection (ESD) for superficial esophageal cancer or suspected superficial esophageal neoplasia 3. Availability of diagnostic endoscopic images obtained before ESD and pre-procedural endoscopic images acquired on the day of ESD 4. Endoscopic image data stored in a format suitable for research use 5. Availability of post-ESD histopathologic assessment allowing confirmation of the final diagnosis and evaluation of invasion depth (M or SM) 6. Patients prospectively and consecutively enrolled during the study period at participating institutions (consecutive enrollment) 7. Patients who voluntarily provide written informed consent for study participation
Exclusion criteria
Exclusion criteria: Exclusion Criteria (Applicable to All Three Participating Centers) 1. Patients without available histopathologic results or for whom assessment of invasion depth is not possible 2. Poor-quality endoscopic images that preclude adequate lesion evaluation (e.g., out-of-focus images, severe motion artifacts, or limited visualization) 3. Cases in which only incomplete image data are available and the target lesion is not adequately included 4. Patients who have received prior treatment that may affect the assessment of invasion depth, including endoscopic resection, surgery, radiotherapy, or chemotherapy 5. Patients with missing key clinical information or image data required for study conduct 6. Patients deemed unsuitable for study participation by the Principal Investigator
Design outcomes
Primary
| Measure | Time frame |
|---|---|
| Accuracy of the AI model in predicting invasion depth (M vs SM) of esophageal cancer | — |
Secondary
| Measure | Time frame |
|---|---|
| Diagnostic performance of the AI model including sensitivity, specificity, and area under the curve (AUC) | — |
Countries
Korea, Republic of
Contacts
Public ContactHyo Soon Yoo
Samsung Medical Center
Outcome results
None listed