Skip to content

Systematic Research on Monitoring, Staging Diagnosis, and Risk Prediction of Chronic Kidney Disease Based on Deep Learning in Magnetic Resonance Imaging

Systematic Research on Monitoring, Staging Diagnosis, and Risk Prediction of Chronic Kidney Disease Based on Deep Learning in Magnetic Resonance Imaging

Status
Recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2500099020
Enrollment
Unknown
Registered
2025-03-17
Start date
2025-03-17
Completion date
Unknown
Last updated
2025-03-24

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

Conditions

Chronic Kidney Disease

Interventions

Training Set (Part I: Retrospective):None
Validation Set (Part I: Retrospective Prospective):None
Training Set (Part II: Retrospective):None
Validation Set (Part II: Retrospective Prospective):None
Training Set (Part III: Retrospective):None
Validation Set (Part III: Retrospective Prospective):None

Sponsors

Shanghai Municipal Hospital of Traditional Chinese Medicine
Lead Sponsor

Eligibility

Sex/Gender
All

Inclusion criteria

Inclusion criteria: Prospective Study Inclusion Criteria: Diagnostic criteria and staging were developed according to the clinical practice guidelines of the Kidney Disease Prognostic Quality Initiative, and patients with CKD were included. Healthy controls had no history of kidney disease or other conditions that cause CKD (e.g., diabetes mellitus, renal artery stenosis, or vascular disease), and eGFR-SCr and eGFR-cys-c >=90 mL/min/1.73 m^2. Database Inclusion Criteria: 1. CKD patients completed non-contrast renal function MRI scans on 3T MRI with good image quality and no obvious distortion and artifacts. The sequences were all coronal, including: T2WI, ASL, T1 plot, T2* plot (BOLD), and quantitative magnetic sensitivity map (QSM). 2. Obtain general clinical data [gender, age, BMI, systolic/diastolic blood pressure (SBP/DBP)] and basic renal function laboratory indicators [eGFR-SCr and eGFRcys-c of modified MDRD, urine protein (Upr), blood ß2 microglobulin, etc.]. Retrospective database: the test indicators included at least 4 items of blood glucose, eGFR-SCr and eGFR-cyc-s, hemoglobin, electrolyte potassium, NGAL, Upr, and blood ß2 microglobulin; Among them, eGFR-SCr and eGFR-cys-c are required, and the inclusion criteria of the same database are the same.

Exclusion criteria

Exclusion criteria: Retrospective exclusion criteria: incomplete MRI scan sequence; MRI image quality cannot be used for automatic renal extraction, dermal and medullary segmentation, and subsequent deep learning; Laboratory markers related to eGFR-SCr and eGFR-cys-c are lacking. Prospective Exclusion Criteria: 1. Acute kidney injury, renal artery stenosis, renal dialysis treatment, large space-occupying lesions of both kidneys (long axis diameter > 20 mm), resection of one nephro, malignant tumor, cardiac insufficiency, liver cirrhosis and severe ascites in patients with CKD. 2. Patients who have used anticoagulant drugs (aspirin, Plavix, etc.) and diuretics within 3 days before MRI scan, and patients who refuse medical intervention or treatment after discovering abnormal kidney function by themselves. 3. Patients with contraindications to MRI (such as metal artifacts in the scanned area, claustrophobia, pregnancy, age less than 18 years or older than 70 years) are also excluded. 4. Less than 4 items in the laboratory tests of blood glucose, eGFR-SCr and eGFR-cyc-s, hemoglobin, electrolyte potassium, NGAL, Upr, and blood ß2 microglobulin; Among them, the deficiency of related eGFR-SCr and eGFR-cys-c was excluded.

Design outcomes

Primary

MeasureTime frame
Susceptibility;Renal blood flow;Oxygen consumption;T1 values;eGFR-SCr;eGFR-cyc-s;Blood glucose;Hemoglobin;Serum potassium;NGAL;Upr;sß2-MG;

Secondary

MeasureTime frame
T2 values;

Countries

China

Contacts

Public ContactYang Shuohui

Shanghai Municipal Hospital of Traditional Chinese Medicine

caddie_yang1980@aliyun.com+86 134 8289 1027

Outcome results

None listed

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