Development of AI Model for Renal Tumor Diagnosis Using CT and Lab Tests

RenJi Hospital1,922 enrolled

Overview

This multi-center retrospective study aims to develop a multimodal artificial intelligence diagnostic model using preoperative contrast-enhanced CT images and routine laboratory parameters from patients with renal tumors. The model is designed to assist clinicians in accurately predicting the pathological subtypes of renal tumors preoperatively, enabling detailed diagnoses and advancing precision medicine.

Study Type

OBSERVATIONAL

Enrollment

1,922

Conditions

Renal Tumors

Eligibility

Sex: ALL

Medical Language ↔ Plain English

Inclusion Criteria: * Underwent renal tumor resection with a complete postoperative pathological report, and the pathological diagnosis is one of the following types: clear cell renal cell carcinoma, papillary renal cell carcinoma, chromophobe renal cell carcinoma, renal angiomyolipoma, or renal oncocytoma. * Complete and available four-phase contrast-enhanced CT scans prior to surgery. * Complete and available routine laboratory test results prior to surgery. Exclusion Criteria: * Incomplete CT data or poor image quality that affects diagnostic analysis. * A time interval of more than three months between imaging or laboratory testing and pathological diagnosis. * Patients diagnosed with fat-rich renal angiomyolipoma (AML). * Pathological diagnosis indicating the coexistence of two or more pathological types of renal tumors.

Locations (1)

Shanghai Jiaotong University School of Medicine, Renji Hospital

Shanghai, Shanghai Municipality, China

Outcomes

Primary Outcomes

postoperative pathological report

Time frame: From the time of surgery to the release of the postoperative pathological report (typically within 2 weeks post-surgery).

Data from ClinicalTrials.gov

This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.