Gastric Cancer, Pancreatic Cancer, Non-Small Cell Lung Cancer, and Head and Neck Squamous Cell Carcinoma
Conditions
Interventions
Gold Standard:Using the pathological results after lymph node dissection as the gold standard
Index test:An AI model for diagnosing regional lymph node metastasis related to specific cancer types (gastric cancer, pancreatic cancer, NSCLC, HNSCC).
Sponsors
Changhai Hospital, The Navy Military Medical University
Eligibility
Sex/Gender
All
Age
18 Years to 90 Years
Inclusion criteria
Inclusion criteria: (1) Age >= 18 years old, gender not restricted. (2) Confirmed as malignant tumor through histopathological examination (or in some cases, highly suspected based on clinical imaging and planned for treatment). The study aims to cover various primary tumor types and will record the specific tumor type during data collection. (3) Received standardized whole-body CT scan at this research center (or cooperative center), and the protocol must include: a) plain scan period; b) at least one contrast-enhanced period (usually the venous phase, with the scan range covering the main lymph node drainage areas from the neck to the pelvis). Arterial phase scans (if applicable) will also be collected. (4) After the CT examination (the recommended time window is within 3 months, which can be adjusted according to clinical practice and recorded), a clear histopathological diagnosis result of at least one lymph node that is visible or locatable on imaging was obtained through lymph node biopsy (puncture, thoracoscopy, laparoscopy, mediastinoscopy, EBUS/EUS-FNA to obtain tissue) or radical surgery/lymph node dissection. (5) The CT image quality is good and meets the requirements for AI analysis and radiologist diagnosis (without severe motion artifacts, metal artifacts, etc., that interfere). (6) The obtained pathological results can be accurately matched and corresponding to the specific lymph nodes on the CT image through surgical records, pathological reports, intraoperative markings, or other reliable methods. (7) (Only for the Phase 2) The patient voluntarily participates in the study and signs a written informed consent form.
Exclusion criteria
Exclusion criteria: (1) The quality of the CT images is poor and cannot meet the requirements for diagnosis or AI analysis. (2) The CT scanning range is incomplete, failing to cover the main lymph node regions, or lacking necessary plain scan/enhancement phases. (3) It is impossible to clearly and reliably locate the lymph nodes corresponding to the pathological results on the CT images (image-pathology mismatch). (4) Only lymph node cytological (such as FNA smear) results are available, but no histopathological (paraffin section) results. (5) The patient has a severe history of contrast agent allergy or other contraindications for CT enhanced scanning (mainly affecting prospective enrollment). (6) (Only for the Phase 2) The patient refuses to sign the informed consent form or is unwilling to follow the research procedures. (7) Participating in other clinical trials that may interfere with the assessment of this study.
Design outcomes
Primary
| Measure | Time frame |
|---|---|
| AUC; | — |
Secondary
| Measure | Time frame |
|---|---|
| sensitivity;specificity;negative predictive value (NPV);positive predictive value (PPV);Accuracy;AUC-PRC; | — |
Countries
China
Contacts
Public ContactYun Bian
Changhai Hospital, The Navy Military Medical University
Outcome results
None listed