Construction of a Preoperative Staging Prediction Model for Cervical Cancer Based on Machine Learning and Deep Learning: A Retrospective Study

Status

Active, not recruiting

Phases

Early Phase 1

Study type

Observational

Source

ChiCTR

Registry ID

ChiCTR2600116794

Enrollment

Unknown

Registered

2026-01-15

Start date

2026-03-05

Completion date

Unknown

Last updated

2026-01-27

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

Conditions

Cervical Cancer?

Interventions

Observational group:None

Eligibility

Sex/Gender

Female

Age

18 Years to 80 Years

Inclusion criteria

Inclusion criteria: 1. Age between 18 and 80 years; 2. Histologically confirmed diagnosis of cervical malignancy; 3. Availability of complete preoperative MRI and/or PET-CT imaging data; 4. No prior neoadjuvant chemotherapy or other anticancer treatments before the MRI/PET-CT examination.

Exclusion criteria

Exclusion criteria: 1. Missing critical pathological data or non-standardized documentation; 2. Lesions too small for accurate delineation; 3. Prior chemotherapy or other treatments that may alter tumor imaging characteristics; 4. Cases involving recurrent disease or re-operation; 5. Poor image quality or absence of essential scanning sequences.

Design outcomes

Primary

Measure	Time frame
Gender, age at diagnosis, tumor location, tumor size, lymph node metastasis, vascular tumor thrombus, perineural invasion, TNM stage, FIGO stage;Molecular pathology: Immunohistochemistry results, genetic testing results;Radiomics features: First-order histogram features, shape features, texture features, Gaussian wavelet transform filters from MR/PET-CT images;Model evaluation metrics: ROC curve (AUC); AIC (Akaike Information Criterion); Calibration plot (goodness-of-fit); C-index;	—

Measure

Time frame

Gender, age at diagnosis, tumor location, tumor size, lymph node metastasis, vascular tumor thrombus, perineural invasion, TNM stage, FIGO stage;Molecular pathology: Immunohistochemistry results, genetic testing results;Radiomics features: First-order histogram features, shape features, texture features, Gaussian wavelet transform filters from MR/PET-CT images;Model evaluation metrics: ROC curve (AUC); AIC (Akaike Information Criterion); Calibration plot (goodness-of-fit); C-index;

—

Countries

China

Contacts

Public ContactLuo Ning

The First Affiliated Hospital of Sun Yat-sen University

luon28@mail.sysu.edu.cn+86 20 8775 5766

Outcome results

None listed

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