Artificial Intelligence-assisted HER2 Expression Assessment in Urothelial Carcinoma Based on Imaging-pathology Omics

This study aims to build upon previous research by using artificial intelligence methods to fuse multimodal data from imaging and pathology to construct a predictive model for HER2 expression in urothelial carcinoma. The model's performance will be validated and optimized using a multicenter cohort study, ultimately achieving accurate and rapid prediction of HER2 expression. This will guide precise decision-making for further HER2-targeted therapy and improve patient prognosis. Big data analysis and deep learning will also assist physicians in more accurately diagnosing the disease and developing personalized treatment plans. The research findings will promote the integration and development of artificial intelligence technology with the healthcare industry in the application of multimodal data from clinical, imaging, and pathology perspectives.

Study Overview

• Status: Enrolling by invitation
• Conditions: Diagnostic; HER2; Urothelial Carcinoma (UC)

Detailed Description

1. Patient Data Collection and Construction of a Multimodal Dataset and Sample Library

   This study will construct a standardized, high-quality multimodal urothelial carcinoma data warehouse. The core is to collect patient data from those pathologically diagnosed with urothelial carcinoma who possess preoperative multiparametric MRI (T1WI, T2WI, DWI/ADC) and paired digital H&E whole slide images (WSI). All cases must use expert-reviewed immunohistochemical results as the gold standard label for HER2 status, and be accompanied by complete clinical data. We will establish strict inclusion and exclusion criteria to ensure data quality, and utilize a professional platform to de-identify, standardize, and correlate clinical, pathological, and imaging data for storage, laying a solid data foundation for subsequent analysis.

2. Extraction and Screening of Imaging/Pathological Features of Urothelial Carcinoma Patients

   This stage aims to extract quantitative features from macroscopic imaging and microscopic pathological images. For MRI, radiologists will manually delineate the three-dimensional region of the tumor (VOI), and then use radiomics tools to extract a large number of quantitative features describing tumor intensity, shape, and texture heterogeneity. For H&E pathological sections, a deep learning/machine learning model was used to automatically segment the tumor region. High-dimensional deep features from millions of image patches were extracted using pre-trained convolutional neural networks, and then aggregated into a feature vector for the entire section using a multi-instance learning framework. Finally, statistical tests and learning methods such as LASSO were combined to select the most relevant and stable feature subset for HER2 status assessment, eliminating redundancy and providing input for model construction.

3. Constructing an AI-Assisted HER2 Assessment and Prediction Model for Urothelial Carcinoma The core task is to integrate the selected multimodal features and construct a high-precision prediction model. We will explore various machine learning algorithms (such as XGBoost and Random Forest) and deep learning architectures, focusing on multimodal fusion strategies: including late-stage fusion that simply splices together radiomics, pathological deep features, and clinical features, and mid-stage fusion (such as using attention mechanisms) that involves feature interaction in the middle layers of the model, to capture deep complementary information between macroscopic imaging and microscopic pathology. The final output of the model is not a simple binary division, but rather provides a continuous probability value (0-1) for HER2 positivity, providing more decision-making basis for clinicians, thereby achieving non-invasive preoperative prediction. (4) Model Validation and Optimization To rigorously evaluate the model's generalization ability and clinical applicability, we performed cross-validation and hyperparameter optimization on an internal dataset, and conducted blinded testing. Evaluation metrics included statistical indicators such as sensitivity, specificity, positive predictive value, and negative predictive value. Simultaneously, interpretability technologies such as Grad-CAM were used to generate heatmaps to locate key imaging areas and pathological morphological features, enhancing physicians' understanding and trust in the model, and potentially uncovering new biological insights.

• Study Type: Observational
• Enrollment (Estimated): 4000

Contacts and Locations

Study Locations

• China
  • Chaoyang District
    • Beijing, Chaoyang District, China, 100021
      • National Cancer Center / Cancer Hospital, Chinese Academy of Medical Sciences Beijing

Participation Criteria

Eligibility Criteria

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

• Sampling Method: Non-Probability Sample

Study Population

We collected imaging and pathological data from patients diagnosed with urothelial carcinoma. Using artificial intelligence, we fused multimodal data from imaging and pathology to construct a predictive model for HER2 expression in urothelial carcinoma. The model's performance was validated and optimized using a multi-center cohort study, ultimately achieving accurate and rapid prediction of HER2 expression. This will guide precise decision-making for further HER2-targeted therapy and improve patient prognosis.

Description

Inclusion Criteria:

• Age ≥ 18 years.
• Patients pathologically diagnosed with urothelial carcinoma.
• Possession of pre-biopsy or pre-operative multiparametric MRI raw data.
• Possession of corresponding paraffin-embedded tissue blocks and digital whole-section images.
• Possession of HER2 status report confirmed by immunohistochemistry.
• Signed informed consent form.

Exclusion Criteria:

• Contraindications to MRI, such as presence of metallic implants or claustrophobia.
• Patients with missing baseline clinical or pathological information.
• Patients who have received neoadjuvant therapy.
• Patients with a history of other malignant tumors.
• Patients with mixed or non-urothelial carcinoma pathology.

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort
Patient diagnosed with urothelial carcinoma by pathology.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Artificial intelligence predicts HER2 expression in urothelial carcinoma	Based on artificial intelligence (AI) technology, this study aims to establish a predictive model by quantitatively mapping the correlation between annotated whole-section images of urothelial carcinoma and MRI scans, identifying common characteristics, and ultimately building a predictive model. Firstly, this model can accurately assess the HER2 status of bladder cancer, eliminating the need for immunohistochemistry to obtain detailed pathological information. Secondly, the established AI predictive model can accurately diagnose the benign or malignant, invasive, grade, and subtype of bladder cancer by predicting the subject's MRI images before biopsy or surgery.	Through study completion, an average of 24 months

Collaborators and Investigators

• Sponsor: Cancer Institute and Hospital, Chinese Academy of Medical Sciences
• Collaborators: Shanxi Province Cancer Hospital; The First Affiliated Hospital of Anhui Medical University; RenJi Hospital; Xinhua Hospital, Shanghai Jiao Tong University School of Medicine; First Hospital of China Medical University; Huaxi Hospital; Huadong Hospital; Cancer Hospital Chinese Academy of Medical Science, Shenzhen Center

Study record dates

Study Major Dates

• Study Start (Estimated): March 2, 2026
• Primary Completion (Estimated): June 3, 2028
• Study Completion (Estimated): June 3, 2030

Study Registration Dates

• First Submitted: March 2, 2026
• First Submitted That Met QC Criteria: March 2, 2026
• First Posted (Actual): March 6, 2026

Study Record Updates

• Last Update Posted (Actual): March 6, 2026
• Last Update Submitted That Met QC Criteria: March 2, 2026
• Last Verified: March 1, 2026

More Information

Terms related to this study

• Keywords: Urothelial Carcinoma; HER2; Artificial Intelligence; Histopathology; Radiomics; Diagnostic Model
• Additional Relevant MeSH Terms: Pathologic Processes; Neoplasms; Neoplasms by Histologic Type; Neoplasms, Glandular and Epithelial; Carcinoma; Pathological Conditions, Signs and Symptoms; Disease; Carcinoma, Transitional Cell
• Other Study ID Numbers: NCC5690

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: UNDECIDED