AI-Guided TURBT for Bladder Cancer (AITURBT)

Evaluation of the Efficacy and Safety of an AI Navigation Planning System in Transurethral Resection of Bladder Tumor: A Prospective, Randomized, Controlled, Single-Center Study

This prospective, randomized, controlled, single-center study evaluates whether an AI navigation planning system used during Transurethral Resection of Bladder Tumor (TURBT) can improve recurrence-free survival in patients with non-muscle invasive bladder cancer. Participants are randomized to receive either standard TURBT surgery or standard TURBT with AI-assisted real-time tumor identification and resection margin guidance. The primary outcome is recurrence-free survival, with secondary outcomes including surgical duration, intraoperative blood loss, perioperative bleeding, and complication rates. The study aims to provide high-level evidence on the clinical utility of AI-assisted surgical navigation in bladder cancer treatment.

Study Overview

• Status: Not yet recruiting
• Conditions: Bladder Cancer
• Intervention / Treatment: Device: AI Navigation Planning System

• Study Type: Interventional
• Enrollment (Estimated): 170
• Phase: Not Applicable

Contacts and Locations

• Study Contact: Name: Lingzhi Du, MD; Phone Number: +86-19839759960; Email: 20301050222@fudan.edu.cn
• Study Contact Backup: Name: Jiajun Wang, MD; Phone Number: +86-15201926887; Email: wang.hang@zs-hospital.sh.cn

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

Description

Inclusion Criteria:

• Age between 18 and 75 years, male or female.
• Clinical or imaging diagnosis of non-muscle invasive bladder cancer (NMIBC) according to international diagnostic standards (CT, MRI, ultrasound, cystoscopy).
• Medically fit for TURBT surgery, with normal or essentially normal renal, cardiopulmonary, and hepatic function.
• Willing and able to provide written informed consent (signed by the patient or a legally authorized representative).

Exclusion Criteria:

• Pregnant or lactating women.
• Severe dysfunction of other vital organs (heart, lungs, kidneys, etc.).
• Imaging evidence of distant metastases.
• Severe infection or active inflammation.
• Psychiatric disorders or inability to comply with study procedures.
• Any other condition deemed by the investigator as inappropriate for participation.

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Treatment
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: Double

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
No Intervention: control; Participants in the control group undergo conventional Transurethral Resection of Bladder Tumor (TURBT) following standard clinical guidelines. The surgeon determines tumor margins and resection range based solely on white-light cystoscopy findings and personal clinical experience, without the use of the AI navigation planning system. All perioperative management, including anesthesia, postoperative care, and follow-up protocols, is identical to that of the experimental group.
Experimental: AI-navigation; Participants in the experimental group undergo standard TURBT with the assistance of the AI navigation planning system. During surgery, the real-time cystoscopic video is processed by the system, which automatically identifies suspicious tumor regions and dynamically displays recommended resection margins as visual overlays on the surgical monitor. The surgeon integrates this AI-generated information with clinical judgment to determine the final resection plan, with the ultimate decision-making authority remaining with the surgeon at all times. All perioperative management, including anesthesia, postoperative care, and follow-up protocols, is identical to that of the control group.	Device: AI Navigation Planning System; The AI Navigation Planning System is a software-based medical device that processes real-time video from a standard cystoscope during TURBT. It uses deep learning algorithms to automatically identify suspicious tumor regions and perform semantic segmentation of tumor boundaries on the endoscopic video stream. The system overlays the identified tumor areas and dynamically recommended resection margins onto the surgical monitor as visual guidance for the surgeon. The system achieves a diagnostic accuracy of 97% and an AUC of 0.97 in multi-center validation. It is designed to assist surgeons in achieving more complete tumor resection by reducing reliance on subjective visual assessment alone. The system operates locally on a standard workstation without requiring internet connectivity, ensuring data security.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Recurrence-Free Survival (RFS)	Time from surgery to the first occurrence of tumor recurrence confirmed by imaging (CT/MRI/ultrasound) or cystoscopy with pathological biopsy, or death from any cause, whichever occurs first. Participants without an event at the end of follow-up will be censored at the date of the last known follow-up visit.	Time Frame: Up to 24 months post-surgery

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Surgical Duration	Time from cystoscope entry into the bladder to successful placement of the urinary catheter at the end of surgery, recorded in minutes.	Day of surgery (intraoperative)
Intraoperative Blood Loss	Total blood loss during surgery, calculated by suction canister fluid minus irrigation fluid plus gauze weighing method, recorded in milliliters (mL).	Day of surgery (intraoperative
Perioperative Bleeding Rate	Proportion of participants experiencing bleeding requiring interventional therapy, unplanned reoperation, or blood transfusion within 14 days post-surgery.	Up to 14 days post-surgery
Number of Lesions Detected Intraoperatively	Total number of tumor lesions identified and documented during the surgical procedure.	Day of surgery (intraoperative)
Postoperative Residual Lesion Rate	Proportion of participants with residual or newly detected lesions at the first postoperative cystoscopy follow-up visit.	Up to 3 months post-surgery
Surgery-Related Complication Rate	Incidence and severity of postoperative complications graded according to the Clavien-Dindo Classification (Grade I to V) within 30 days post-surgery.	Up to 30 days post-surgery

Collaborators and Investigators

• Sponsor: Fudan University
• Investigators: Study Director: Jiajun Wang, MD, Fudan University

Study record dates

Study Major Dates

• Study Start (Estimated): August 1, 2026
• Primary Completion (Estimated): May 31, 2027
• Study Completion (Estimated): November 30, 2028

Study Registration Dates

• First Submitted: July 19, 2026
• First Submitted That Met QC Criteria: July 23, 2026
• First Posted (Actual): July 27, 2026

Study Record Updates

• Last Update Posted (Actual): July 27, 2026
• Last Update Submitted That Met QC Criteria: July 23, 2026
• Last Verified: July 1, 2026

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Urogenital Diseases; Urogenital Neoplasms; Neoplasms by Site; Neoplasms; Male Urogenital Diseases; Urologic Diseases; Female Urogenital Diseases; Female Urogenital Diseases and Pregnancy Complications; Urologic Neoplasms; Urinary Bladder Diseases; Urinary Bladder Neoplasms
• Other Study ID Numbers: ZS-AI based navigation

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO
• IPD Plan Description: IPD will not be shared due to: (1) the informed consent form does not include provisions for sharing individual participant data with external researchers; (2) intraoperative video data contain sensitive patient information that cannot be fully anonymized; (3) institutional data governance policies restrict data sharing beyond the primary study team; and (4) the AI navigation system involves proprietary technology, and the validation dataset requires controlled access to protect intellectual property. Summary-level results will be made available in peer-reviewed publications.

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No