Skip to content

A Prospective Study for Breast Core Needle Biopsy Pathological Diagnosis with a Whole-slide Level Multimodal Vision-Language Model

A Prospective Study for Breast Core Needle Biopsy Pathological Diagnosis with a Whole-slide Level Multimodal Vision-Language Model

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2500111486
Enrollment
Unknown
Registered
2025-10-31
Start date
2025-11-01
Completion date
Unknown
Last updated
2026-08-10

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

breast tumor

Interventions

Gold Standard:The reference standard for calculating diagnostic accuracy in this study was the final diagnosis for each breast core needle biopsy slide. This was established through an independent rev
Index test:multimodal vision-language model MINDx

Sponsors

Fudan University Shanghai Cancer Center
Lead Sponsor

Eligibility

Sex/Gender
All
Age
No minimum to 100 Years

Inclusion criteria

Inclusion criteria: Inclusion Criteria: - All cases that underwent breast core needle biopsy and successfully obtained a final pathological diagnosis in FUSCC. There are no restrictions on external conditions such as age, gender, and type of lesion.

Exclusion criteria

Exclusion criteria: 1. Cases undergoing anti-tumor drug treatment before core needle biopsy, where the purpose of the biopsy is efficacy evaluation; 2. Cases with severely degenerated specimens and insufficient tissue available for diagnosis; 3. Cases with poor quality of specimen sections or scanned images, including the following specific situations: a. Tissue folds or damage accounting for more than 50% of the Region of Interest (ROI) in the section; b. Annotation handwriting or dye covering more than 50% of the ROI area; c. Severely blurred or out-of-focus scanned images that affect the identification of pathological structures.

Design outcomes

Secondary

MeasureTime frame
Macro-F1 Score;Diagnostic Accuracy for Malignant Tumors;Pairwise preference by senior pathologists;Satisfaction score towards the MINDx assistant system;Proportion of cases requiring additional immunohistochemical testing for diagnosis;Average diagnostic time per case;

Primary

MeasureTime frame
Diagnostic Accuracy;

Countries

China

Contacts

Public ContactYang Wentao

Fudan University Shanghai Cancer Center

yangwt2000@163.com+86 21 6417 5590

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Aug 25, 2026