AI-Based Imaging Model for Bladder Cancer Prediction

Sun Yat-sen University

Status: Enrolling

Conditions: Develop a CT-based Tumor Budding Predictive Model for Bladder Cancer Using Deep Learning Algorithms

Treatments: Other: No specific interventions.

Study type: Observational

Funder types: Other

Identifiers

NCT06442839

SL-II2023-303-02

Details and patient eligibility

About

Bladder cancer is the ninth most common malignant tumor worldwide, characterized by high malignancy and poor prognosis. We intend to develop a CT-based tumor budding predictive model for bladder cancer using deep learning algorithms. This model will facilitate preoperative assessment of patient conditions, enabling the formulation of more precise and personalized treatment plans.

Enrollment: 2,000 estimated patients

Sex: All

Ages: 18+ years old

Volunteers: No Healthy Volunteers

Inclusion criteria

1. Bladder cancer patients treated from January 1, 2014, to January 1, 2023;
2. Hospitalized and underwent transurethral resection of bladder tumor (TURBT) or radical cystectomy;
3. Complete clinical, preoperative CT, and pathological data.

Exclusion criteria

1. Patients who previously underwent surgical treatment for bladder cancer at other centers, making it difficult to obtain their preoperative data;
2. Patients with other concurrent pelvic or urinary system malignancies;
3. Patients with poor quality, low resolution, or faded CT or pathological images.

Trial design

2,000 participants in 6 patient groups

training cohort

Description:

for training the model, from Sun Yat-sen Memorial Hospital of Sun Yat-sen University.

Treatment:

Other: No specific interventions.

internal validation cohort

Description:

used to evaluate the model's performance, is from Sun Yat-sen Memorial Hospital of Sun Yat-sen University.

Treatment:

Other: No specific interventions.

external validation cohort 1

Description:

used to evaluate the model's performance, is from the Third Affiliated Hospital of Sun Yat-sen University.

Treatment:

Other: No specific interventions.

external validation cohort 2

Description:

used to evaluate the model's performance, is from the Second Affiliated Hospital of Dalian Medical University.

Treatment:

Other: No specific interventions.

external validation cohort 3

Description:

used to evaluate the model's performance, is from the First Affiliated Hospital of Chongqing Medical University.

Treatment:

Other: No specific interventions.

external validation cohort 4

Description:

used to evaluate the model's performance, is from the Yan'an Hospital Affiliated to Kunming Medical University.

Treatment:

Other: No specific interventions.