Skip to content

Key technology research on constructing a breast NME lesion risk stratification DL-CIR model based on deep learning

Key technology research on constructing a breast NME lesion risk stratification DL-CIR model based on deep learning

Status
Recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2500106039
Enrollment
Unknown
Registered
2025-07-16
Start date
2024-07-15
Completion date
Unknown
Last updated
2025-07-21

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

Conditions

Breast Lesions, breast cancer

Interventions

Gold Standard:Pathological results of breast biopsy or surgery.
Index test:Clinical models, radiomics models, deep learning models, and DL-CIR models constructed based on training set data.

Sponsors

Tongde Hospital of Zhejiang Province
Lead Sponsor

Eligibility

Sex/Gender
Female
Age
18 Years to 90 Years

Inclusion criteria

Inclusion criteria: 1.Displayed as NME on breast DCE-MRI image; 2.If there is complete biopsy or surgical pathology data, and MRI results show disappearance or stability of the lesion during a 2-year follow-up without biopsy or surgical resection, it can be considered benign;

Exclusion criteria

Exclusion criteria: 1.Breast DCE-MRI image shows mass or punctate enhancement; 2.Prior to breast MRI examination, lesion biopsy and breast radiotherapy and chemotherapy were performed; 3.Poor quality of breast MRI images affects image analysis;

Design outcomes

Primary

MeasureTime frame
Deep Learning Features;ROC curve;AUC value;

Secondary

MeasureTime frame
imaging features;radiomics features;

Countries

China

Contacts

Public ContactGuoqun Mao

Tongde Hospital of Zhejiang Province

maoguoqun123@163.com+86 571 89972226

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Feb 4, 2026