Prospective Real-World Study of Multimodal AI

NCT07269236 | Enrolling By Invitation

• 1 other identifier: NFEC-2025-602
• Study Type: observational
• Target: 2,000
• Locations: 1 country
• Sites: 2

Brief Summary

This project proposes to collect prospective multimodal data-such as pathology, imaging, and clinical information-and to perform integrative analyses. AI technologies can offer novel solutions for disease classification, tumor grading, histological subtyping, molecular subtyping, selection of chemotherapy regimens, risk stratification, treatment response prediction, report generation, and intelligent question-answering. This research provides important support for precision medicine and individualized treatment and has significant theoretical and practical implications. Conducting a prospective randomized controlled study better aligns with clinical application requirements and can accelerate the comprehensive deployment of AI systems.

Conditions

Pancancer

Outcome Measures

Primary Outcomes (1)

• Area under ROC curve (AUC)

  Area under the curve

  Diagnostic evaluation will be performed within 1 week when the WSIs or CT are obtained

Secondary Outcomes (2)

• Specificity

  Diagnostic evaluation will be performed within 1 week when the WSIs or CT are obtained

• Sensitivity

  Diagnostic evaluation will be performed within 1 week when the WSIs or CT are obtained

Study Arms (3)

ZSCPH validation dataset

Approximately 1,000 prospective patients across various disease types will have digitized scanned images, imaging information, and clinicopathological data provided, collected from Zhongshan People's Hospital (Zhongshan, Guangdong, China) between October 2025 and September 2030.

NFHSMU validation dataset

We conducted a prospective validation study to compare the diagnostic performance among pathologists, our multimodal artificial intelligence model . This study was initiated on October 1, 2025 at Nanfang Hospital, Southern Medical University (NFHSMU).

Randomized controlled trial

We conducted a prospective randomized controlled trial to compare the diagnostic performance among pathologists, the multimodal artificial intelligence model, and pathologist-with-model-assisted diagnosis at Nanfang Hospital of Southern Medical University (NFHSMU).The trial commenced data collection on October 1, 2025 to establish the NFHSMU randomized controlled trial dataset. Following quality control, the slides were randomly allocated (1:1:1 ratio) into three groups:Model-only group/Pathologist-only group/Model-assisted pathologist group.

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Healthy Volunteers: No
• Age Groups: Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

Inclusion criteria comprised patients with definitive pathological diagnoses. All cases were prospectively collected from Nanfang Hospital, Southern Medical University (NFHSMU) .

You may qualify if:

• Age 18 years or older.
• Patients with available digitized pathology slides, radiological imaging, and corresponding clinical data.

You may not qualify if:

• Patients with missing data or specimens not meeting quality control requirements for analysis.

Sponsors & Collaborators

• Nanfang Hospital, Southern Medical University: lead

Study Sites (2)

Nanfang Hospital, Southern Medical University

Guangzhou, Guangdong, 510515, China

Location

Zhongshan City People's Hospital

Zhongshan, Guangdong, 513000, China

Location

Biospecimen

Retention: SAMPLES WITHOUT DNA

Samples Without DNA: Samples retained, with no potential for DNA extraction from any retained samples (e.g., fixed tissue, plasma)

Study Officials

• Li Liang

  Nanfang Hospital, Southern Medical University

  STUDY DIRECTOR

Study Design

• Study Type: observational
• Observational Model: COHORT
• Time Perspective: PROSPECTIVE
• Sponsor Type: OTHER
• Responsible Party: SPONSOR

Study Record Dates

• First Submitted: November 25, 2025
• First Posted: December 8, 2025
• Study Start: October 1, 2025
• Primary Completion (Estimated): September 1, 2030
• Study Completion (Estimated): September 1, 2030
• Last Updated: December 8, 2025

Record last verified: 2025-11

Data Sharing

• IPD Sharing: Will not share

Locations

CN (2)