Skip to content

Construction and research of a non-invasive risk prediction model for peripancreatic nerve invasion in pancreatic cancer based on Artificial Intelligence

Construction and research of a non-invasive risk prediction model for peripancreatic nerve invasion in pancreatic cancer based on Artificial Intelligence

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2500104673
Enrollment
Unknown
Registered
2025-06-20
Start date
2025-06-23
Completion date
Unknown
Last updated
2025-06-24

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Pancreatic cancer

Interventions

Gold Standard:The postoperative pathological diagnosis was pancreatic ductal adenocarcinoma
Index test:3D spatial contact characteristics of vascular-tumors based on deep learning algorithms

Sponsors

Changhai Hospital, The Navy Military Medical University
Lead Sponsor

Eligibility

Sex/Gender
All

Inclusion criteria

Inclusion criteria: 1. Histologically confirmed diagnosis of PDAC; 2. Available preoperative contrast-enhanced CT scans with <=2 mm slice thickness, acquired within 30 days prior to surgery; 3. Underwent R0 or R1 surgical resection at Changhai Hospital between 2015 and 2019; 4. Documented pathological assessment of EPNI status; 5. Complete clinicopathological data available for model training and validation.

Exclusion criteria

Exclusion criteria: 1. Diagnosis of non-pancreatic primary tumors; 2. Presence of distant metastasis at the time of diagnosis; 3. CT image quality insufficient for segmentation (e.g., severe motion artifacts, slice thickness >2 mm); 4. Incomplete clinical, imaging, or histopathological records; 5. Cases involving synchronous malignancies or prior oncological treatments, except for patients receiving neoadjuvant therapy (who were included as a separate validation subgroup).

Design outcomes

Primary

MeasureTime frame
Area Under the Receiver Operating Characteristic Curve, AUC;Sensitivity;Specificity;

Secondary

MeasureTime frame
Negative predictive value (NPV);Positive predictive value (PPV);

Countries

China

Contacts

Public ContactYun Bian

Changhai Hospital, The Navy Military Medical University

bianyun2012@foxmail.com+86 138 1635 7024

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Feb 4, 2026