An MRI-Based Study of Intelligent Pathological Subtyping and Grading of Renal Tumors

Overview

This retrospective + prospective, non-interventional study aims to develop and evaluate artificial intelligence methods for the detection, pathological subtyping, and histological grading of renal tumors using magnetic resonance imaging (MRI). Approximately 900 adult patients with available preoperative renal MRI examinations and postoperative pathological results will be included. The pathological findings will be used as the reference standard for model development and evaluation. In addition to MRI data, selected demographic, clinical, and laboratory information may be incorporated to improve model performance. The study will not change participants' diagnosis, treatment, or follow-up, and no additional examinations or interventions will be required. All study data will be de-identified before analysis. The ultimate goal is to develop an MRI-based intelligent diagnostic approach that may assist clinicians in the preoperative assessment and individualized management of patients with renal tumors.

Detailed description

Renal tumors include multiple benign and malignant pathological subtypes with substantial differences in biological behavior, treatment strategy, and prognosis. Surgical planning and clinical management are closely related to the pathological subtype and histological grade of the tumor. However, accurately determining these pathological characteristics before surgery using conventional MRI interpretation remains challenging.

This is a retrospective + prospective, observational, and non-interventional study. Adult patients with renal tumors will be identified from existing clinical records. Eligible patients will have available renal MRI examinations and corresponding pathological diagnoses, including pathological subtype and, when applicable, histological grade. Cases with unreadable MRI data or images of insufficient quality for analysis will be excluded.

Existing study data will include multisequence MRI examinations, such as T1-weighted imaging, T2-weighted imaging, diffusion-weighted imaging, apparent diffusion coefficient imaging, fat-suppressed imaging, and contrast-enhanced imaging, when available. Demographic information, relevant clinical history, laboratory results, and radiology report information may also be collected. Pathological findings will serve as the reference standard for model training and evaluation. All data will be de-identified before processing and analysis.

The study will develop artificial intelligence models for the following tasks:

1. Detection and localization of renal tumors on multisequence MRI. 2. Segmentation of renal tumors and extraction of quantitative imaging features. 3. Classification of common benign and malignant renal tumor subtypes. 4. Identification of rare pathological subtypes using small-sample or cross-modal learning methods. 5. Prediction of histological grade for malignant renal tumors. 6. Integration of MRI, demographic, clinical, and laboratory information to improve pathological subtyping and grading.

The dataset will be divided into model-development and validation datasets. Additional cases collected from different time periods or participating sources may be used for independent testing. Model performance will be evaluated by comparing artificial intelligence predictions with pathological diagnoses.

This study does not assign any treatment or diagnostic intervention. It will not affect participants' routine clinical care and will not require additional imaging examinations, blood collection, surgery, medication, or follow-up visits. The study is intended to develop an intelligent MRI-based diagnostic system that may support preoperative decision-making for patients with renal tumors.

Interventions

• Diagnostic test MRI-Based Artificial Intelligence Analysis
  Existing preoperative multisequence renal MRI images, including T1-weighted imaging, T2-weighted imaging, diffusion-weighted imaging, apparent diffusion coefficient imaging, fat-suppressed imaging, and contrast-enhanced imaging when available, were retrospectively analyzed using artificial intelligence and deep learning methods. The models were developed to detect and segment renal tumors and to predict pathological subtype and histological grade. Postoperative pathological findings were used as

Primary outcome measures

• Accuracy of MRI-Based Artificial Intelligence for Pathological Subtyping of Renal Tumors [Time frame: At completion of model evaluation using retrospective data collected from January 2021 through July 2026 and prospective data collected through December 2026.]
• Accuracy of MRI-Based Artificial Intelligence for Histological Grading of Malignant Renal Tumors [Time frame: At completion of model evaluation using retrospective data collected from January 2021 through July 2026 and prospective data collected through December 2026.]

Secondary outcome measures (2)

• Performance of the Artificial Intelligence Model for Renal Tumor Detection [Time frame: At completion of model evaluation using retrospective data collected from January 2021 through July 2026 and prospective data collected through December 2026.]
• Accuracy of Artificial Intelligence-Based Renal Tumor Segmentation [Time frame: At completion of model evaluation using retrospective data collected from January 2021 through July 2026 and prospective data collected through December 2026.]

Eligibility criteria

Inclusion criteria

• Patients aged 18 years or older.
• Patients diagnosed with a renal tumor.
• Availability of preoperative renal magnetic resonance imaging examinations.
• Availability of a corresponding pathological diagnosis, including pathological subtype and, where applicable, histological grade.
• Magnetic resonance images that can be successfully retrieved and are of - - sufficient quality for image analysis.

Exclusion criteria

• Absence of renal magnetic resonance imaging data.
• Absence of a corresponding pathological diagnosis or insufficient pathological subtype or grading information.
• Magnetic resonance images that cannot be retrieved, opened, or read.
• Poor image quality that precludes reliable image annotation or artificial intelligence analysis.

Criteria are shown verbatim from the registry (in English). Final eligibility is always assessed by the study center.

Healthy volunteers: No

Study design

Observational model

Case-only

Study locations

China · 1 center

• Cancer Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College — Beijing

Identifiers

NCT: NCT07743749 · NCC-020738

Primary sources (government registries)

• ClinicalTrials.gov (US NIH)

View this study on ClinicalTrials.gov ↗