Integrating an AI-Driven Hydronephrosis Decision-Making Tool

NCT07581223 | Not Yet Recruiting DMC

• 1 other identifier: 3474
• Study Type: interventional
• Target: 322
• Locations: 0 countries
• Sites: N/A

Brief Summary

Hydronephrosis is a common congenital kidney anomaly. While most cases resolve on their own, some require surgery. Clinicians rely on repeated ultrasounds and sometimes invasive tests to decide if surgery is needed, but predicting outcomes is difficult. Researchers at SickKids developed an AI model that analyzes ultrasound images to assist in diagnosing and managing hydronephrosis. This study tests how well the AI integrates into real-world care. Clinicians will first make care decisions without AI and then review the AI's prediction before deciding whether to change their plan. A separate expert, unaware of whether AI influenced the first clinician's plan, will make the final decision to ensure care remains unchanged. The study will assess whether AI improves decision-making, reduces unnecessary tests, and fits into clinical workflows. If successful, the AI model could serve as a complementary tool to make diagnoses more efficient and precise while minimizing invasive procedures.

Conditions

Hydronephrosis Congenital; Hydronephrosis

Outcome Measures

Primary Outcomes (1)

• Change in Clinician Management Decisions Following Exposure to the AI Model

  The proportion of clinician management decisions revised immediately after exposure to the AI model output. Management decisions include: (1) discharge, (2) monitor with ultrasound, (3) additional invasive testing, or (4) referral for surgery.

  Immediately after AI model exposure during each case review session, through study completion (average of 6 months)

Secondary Outcomes (2)

• Agreement Between Clinician Decisions and Expert Reference Decisions Using Cohen's Kappa

  Immediately after clinician review and AI model exposure during each case review session, through study completion (average of 6 months)

• Proportion of Management Decision Changes Stratified by Clinician Experience Level

  Immediately after AI model exposure during each case review session, through study completion (average of 6 months)

Study Arms (1)

AI Model Intervention Arm

EXPERIMENTAL

When children with HN are seen in clinic, their ultrasound imaging and history will be provided to an initial clinician who will first formulate a plan of care without access to the AI model as per the standard of care. After the initial plan is documented and before discussion with the primary provider, the initial clinician will then be granted access to the AI model, where they can input the ultrasound images and receive the model's prediction. The clinician can choose to modify or maintain their drafted plan based on the model's output. The clinician's final drafted plan will subsequently be discussed with the blinded final clinical expert (primary provider) who will make the final decision to maintain the standard of care for each patient. The final clinical expert will be blinded to whether the initial clinician changed their plan or not given the AI model

Other: Machine learning model

Interventions

Machine learning model OTHER

The AI intervention is a deep learning algorithm used to predict obstructive hydronephrosis. It was developed at SickKids and has recently completed the silent trial phase. This clinical trial aims to validate the model's clinical integration by assessing its impact on clinician decision-making and care plan recommendations. To uphold standard care, a blinded clinician will make final decisions.

AI Model Intervention Arm

Eligibility Criteria

• Age: 0 Months - 24 Months
• Sex: all
• Healthy Volunteers: No
• Age Groups: Child (0-17)

You may qualify if:

• Seen for HN in-person in the Pediatric Urology clinic with ultrasound scans taken at SickKids
• New and follow-up patients 0-24 months.

You may not qualify if:

• Older than 24m
• Concurrent urinary tract anomalies (duplex configurations; PUV etc.)
• History of renal surgical intervention (post-op patients)

Sponsors & Collaborators

• The Hospital for Sick Children: lead

MeSH Terms

Conditions

Hydronephrosis

Condition Hierarchy (Ancestors)

Kidney Diseases; Urologic Diseases; Female Urogenital Diseases; Female Urogenital Diseases and Pregnancy Complications; Urogenital Diseases; Male Urogenital Diseases

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: NA
• Masking: NONE
• Masking Details: This study will be a partially blinded trial as a third blinded clinician who makes the final clinical decision will be unaware if and what changes were made after clinician exposure to the AI model. Since, standard of care is maintained, patients will not be aware of the impact of the AI model
• Purpose: OTHER
• Intervention Model: SINGLE GROUP
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Nurse Practitioner

Model Details: The intervention is an AI algorithm for the prediction of obstructive HN. When children with HN are seen in clinic, their ultrasound imaging and history will be provided to an initial clinician who will first formulate a plan of care without access to the AI model as per the standard of care. After the initial plan is documented and before discussion with the primary provider, the initial clinician will then be granted access to the AI model, where they can input the ultrasound images and receive the model's prediction. The clinician can choose to modify or maintain their drafted plan based on the model's output. The clinician's final drafted plan will subsequently be discussed with the blinded final clinical expert (primary provider) who will make the final decision to maintain the standard of care for each patient. The final clinical expert will be blinded to whether the initial clinician changed their plan or not given the AI model.

Study Record Dates

• First Submitted: April 16, 2026
• First Posted: May 12, 2026
• Study Start (Estimated): August 1, 2026
• Primary Completion (Estimated): January 31, 2027
• Study Completion (Estimated): January 31, 2027
• Last Updated: May 12, 2026

Record last verified: 2026-05