Skip to content

Diagnostic Performance of Deep Learning Based on Magnetic Resonance Imaging for Predicting Extraprostatic Extension in Prostate Cancer

Diagnostic Performance of Deep Learning Based on Magnetic Resonance Imaging for Predicting Extraprostatic Extension in Prostate Cancer

Status
Active, not recruiting
Phases
Unknown
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2500105361
Enrollment
Unknown
Registered
2025-07-02
Start date
2025-07-02
Completion date
Unknown
Last updated
2025-07-07

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

Conditions

Prostate cancer

Interventions

Gold Standard:Pathological diagnosis
Index test:The Deep Learning Model Based on Magnetic Resonance Imaging

Sponsors

The First Affiliated Hospital of Soochow University
Lead Sponsor

Eligibility

Sex/Gender
Male
Age
45 Years to 90 Years

Inclusion criteria

Inclusion criteria: Inclusion Criteria: Patients diagnosed with prostate cancer by biopsy between January 2015 and 2024 at our hospital and collaborating institutions, with: Available high-quality preoperative 3.0T MRI imaging; Underwent radical prostatectomy; Postoperative pathology clearly documenting the presence or absence of extraprostatic extension (EPE).

Exclusion criteria

Exclusion criteria: Patients with prostate cancer at our hospital or collaborating institutions who did not undergo a clearly documented preoperative 3.0T MRI examination. Patients who received neoadjuvant therapy or had a history of prior local prostate surgery. Patients with missing key clinical data, including PSA results, biopsy pathology, or other essential clinical information. Patients whose MRI data were of poor quality and therefore unsuitable for deep learning modeling or further analysis. Patients whose radical prostatectomy pathology reports did not clearly document the status of extraprostatic extension (EPE).

Design outcomes

Primary

MeasureTime frame
AUC of AI model for predicting extracapsular invasion of prostate cancer;Accuracy;Sensitivity;Specificity;

Countries

China

Contacts

Public Contactzhixin ling

department of urology, the first affiliated hospital of soochow university

lingzhixin@suda.edu.cn+86 159 9544 5546

Outcome results

None listed

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