AI for Newborn Metabolic Screening

NCT07368504 | Not Yet Recruiting

• 1 other identifier: 2025-IRB-0550-P-01
• Study Type: interventional
• Target: 200,000
• Locations: 1 country
• Sites: 1

Brief Summary

The goal of this clinical trial is to evaluate whether an artificial intelligence (AI)-based interpretation system can accurately diagnose inherited metabolic disorders in newborns undergoing routine screening. The main questions it aims to answer are: What is the sensitivity and specificity of the AI system compared to standard manual interpretation? Does the AI system reduce variability in screening results? Researchers will compare the AI interpretation results with those from standard manual review by trained laboratory staff to assess diagnostic performance. Participants will: Have their routine newborn screening blood samples analyzed using both the AI system and standard manual interpretation Be followed according to national newborn screening guidelines if either method indicates a positive result

Conditions

Inherited Metabolic Disorders

Outcome Measures

Primary Outcomes (2)

• Sensitivity of the AI interpretation system for detecting inherited metabolic disorders

  Within 12 months after newborn screening

• Specificity of the AI interpretation system for detecting inherited metabolic disorders

  Within 12 months after newborn screening

Study Arms (1)

AI and Manual Interpretation of Newborn Screening Data

OTHER

Diagnostic Test: Artificial intelligence-based interpretation system for newborn genetic metabolic disease screening

Interventions

Artificial intelligence-based interpretation system for newborn genetic metabolic disease screening DIAGNOSTIC_TEST

This intervention is a deep learning-based software algorithm designed specifically for the interpretation of tandem mass spectrometry (MS/MS) data from routine newborn screening in Chinese neonates. It integrates clinical covariates-including gestational age, birth weight, and blood collection time-to perform multiple-of-the-median (MOM) normalization and simultaneously evaluates 42 inherited metabolic disorders. Unlike existing AI tools developed for older-generation screening panels (e.g., those covering only 29 analytes), this system is trained and validated on over 300,000 real-world Chinese newborn samples, making it the first AI diagnostic tool tailored to China's current expanded newborn screening program.

AI and Manual Interpretation of Newborn Screening Data

Eligibility Criteria

• Age: 2 Days - 28 Days
• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Child (0-17)

You may qualify if:

• Newborns who underwent routine newborn screening for inherited metabolic disorders at the Zhejiang Provincial Newborn Screening Center between May 2025 and December 2027
• Blood samples collected between 2 and 28 days of age
• Availability of complete newborn screening test data and essential clinical information

You may not qualify if:

• Missing, incomplete, or poor-quality screening data
• Duplicate samples from the same newborn

Sponsors & Collaborators

• The Children's Hospital of Zhejiang University School of Medicine: lead

Study Sites (1)

The Children's Hospital, Zhejiang University School of Medicine

Hangzhou, Zhejiang, 310000, China

Location

Central Study Contacts

Hu, PhD

CONTACT

+86-571-86670459; hzz22980825@zju.edu.cn

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: NA
• Masking: NONE
• Purpose: SCREENING
• Intervention Model: SINGLE GROUP
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Principal Investigator

Study Record Dates

• First Submitted: January 16, 2026
• First Posted: January 26, 2026
• Study Start (Estimated): January 1, 2027
• Primary Completion (Estimated): June 30, 2028
• Study Completion (Estimated): November 30, 2028
• Last Updated: January 27, 2026

Record last verified: 2026-01

Locations

CN (1)