Research and Validation of a Big Data-Driven Intelligent Decision-Making System for Hemodialysis

NCT07466329 | Completed

• 1 other identifier: KY2025-1424
• Study Type: observational
• Target: 778
• Locations: 1 country
• Sites: 1

Brief Summary

Objectives and Scope:This observational study aims to leverage real-world data from Huashan Hospital to develop an AI-driven intelligent decision-making system for assessing dialysis adequacy in maintenance hemodialysis (MHD) patients, and to analyze early warning factors contributing to inadequate dialysis. Core Research Question:Can an AI-based early warning and diagnostic model, built on multidimensional big data, identify the risk of inadequate hemodialysis at an ultra-early stage and accurately diagnose composite complications such as cardiovascular and cerebrovascular diseases? Methodology:The study will conduct a retrospective analysis of adult MHD patients treated at Huashan Hospital between January 2011 and September 2025. The dataset encompasses multidimensional variables, including sociodemographics, treatment parameters, laboratory indicators, metabolomics, and physical functions. Utilizing Dynamic Network Biomarkers (DNB) technology to screen for early warning markers, combined with artificial intelligence algorithms such as Neural Networks and Support Vector Machines (SVM), the study will construct two primary models: "Ultra-early Warning" and "Disease State Diagnosis." These models are designed to provide clinical decision support for precise interventions.

Conditions

ESRD (End Stage Renal Disease)

Keywords

End-Stage Renal Disease; Maintenance Hemodialysis

Outcome Measures

Primary Outcomes (1)

• Cardiovascular and Cerebrovascular Diseases (CCVD)

  Clinicians diagnose these conditions based on the American Heart Association (AHA) professional guidelines and diagnostic criteria.

  15 years (From January 2011 to September 2025)

Secondary Outcomes (1)

• Composite Complications

  15 years (From January 2011 to September 2025)

Study Arms (1)

Patients undergoing hemodialysis at this center from January 2011 to September 2025.

Information will be retrospectively collected from patients who underwent hemodialysis at this center between January 2011 and September 2025. Data are primarily sourced from electronic information systems, including the Hemodialysis Electronic Management System, the hospital Health Information System (HIS), and the Inpatient Medical Record System. The dataset encompasses personal information, laboratory results, diagnostic data, medical orders, and nutritional status.

Eligibility Criteria

• Age: 18 Years - 90 Years
• Sex: all
• Healthy Volunteers: No
• Age Groups: Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

The study population consists of patients receiving maintenance hemodialysis at the Hemodialysis Center of Huashan Hospital between January 2011 and September 2025. Eligibility criteria require patients to be aged 18-90 years with a dialysis vintage of at least 3 months. Additionally, participants must have comprehensive clinical and follow-up data maintained within the center's electronic medical record system.

You may qualify if:

• Patients undergoing long-term maintenance hemodialysis with a dialysis vintage of at least 3 months.
• Aged between 18 and 90 years.
• Possess relatively comprehensive hemodialysis records maintained within this center.

You may not qualify if:

• Patients with significantly incomplete dialysis-related data.
• Patients with poor compliance during dialysis or those receiving palliative dialysis.
• Other conditions deemed unsuitable by the investigator.

Sponsors & Collaborators

• Huashan Hospital: lead

Study Sites (1)

Huashan Hospital, Fudan University

Shanghai, Shanghai Municipality, 200040, China

Location

Biospecimen

Retention: SAMPLES WITH DNA

Samples will be collected at three time points: at baseline, at the time of clinical diagnosis of inadequate dialysis, and at the time of clinical diagnosis of dialysis adequacy. For each collection, 1 ml of blood and 1 ml of urine will be obtained to extract and analyze metabolites, including amino acids, carbohydrates, lipids, and nucleotides.

MeSH Terms

Conditions

Renal Insufficiency, Chronic; Kidney Failure, Chronic

Condition Hierarchy (Ancestors)

Renal Insufficiency; Kidney Diseases; Urologic Diseases; Female Urogenital Diseases; Female Urogenital Diseases and Pregnancy Complications; Urogenital Diseases; Male Urogenital Diseases; Chronic Disease; Disease Attributes; Pathologic Processes; Pathological Conditions, Signs and Symptoms

Study Officials

• Jing Chen

  Huashan Hospital

  PRINCIPAL INVESTIGATOR

Study Design

• Study Type: observational
• Observational Model: COHORT
• Time Perspective: RETROSPECTIVE
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Professor

Study Record Dates

• First Submitted: March 8, 2026
• First Posted: March 12, 2026
• Study Start: January 1, 2011
• Primary Completion: September 30, 2025
• Study Completion: September 30, 2025
• Last Updated: March 17, 2026

Record last verified: 2026-01

Data Sharing

• IPD Sharing: Will not share

Locations

CN (1)