Using AI on prostate MRI scans to predict serious prostate cancer
Studies AI models that read prostate MRI scans to predict serious prostate cancer.
Plain-language title generated from the registry record
Registry title: The Development and Validation of MRI-AI-based Predictive Models for csPCa
Key facts
- Where
- Beijing, ChinaSee the site
AI summary: who the study is looking for
From the trial's eligibility criteria. Machine-read from the registry text on this page, so it can be wrong. The registry text below and the trial team decide who can take part.
The trial is looking for men who had a prostate MRI scan and then a prostate biopsy within 3 months, with complete data. It uses the data to build prediction models for clinically-significant prostate cancer.
- Sex
- Men only.
- Earlier treatment
- People cannot join if they had any treatment for prostate cancer before the MRI or biopsy. This includes surgery to remove the prostate, radiotherapy, chemotherapy and hormone therapy.
Main reasons for not taking part
- PSA less than 50ng/ml
- Any prostate cancer treatment before the MRI or biopsy
- Past surgery or 5α-reductase inhibitor treatment for benign prostatic hyperplasia
- MRI done with a urinary catheter or suprapubic catheter in place
- MRI images of poor quality
Read by machine on 12 November 2026.
Where in China
- Peking University First HospitalBeijing, ChinaContact: Yi LIUTaking people
Questions to ask the study team
- How often do I need to come in, and for how long each time?
- Are travel costs paid?
- Can I keep my own doctor?
- What happens if I want to stop?
Contact the study team: see the registry record → Contacts and Locations
Print a page for my doctorTrial details
- Who runs it
- Peking University First Hospital
- Trial code
- NCT06842264
- Record updated
- 29 January 2026
- Original description
This study retrospectively included patients who underwent prostate magnetic resonance imaging (MRI) and subsequent ultrasound-guided prostate biopsy at Peking University First Hospital from January 2019 to December 2023, and prospectively enrolls patients from January 2024 to December 2029. Clinical information such as age, PSA levels, PI-RADS scores, and digital rectal examination findings are collected. A well-performing artificial intelligence model is employed to measure prostate volume, transitional zone volume, and lesion volume using MRI images. Furthermore, prostate-specific antigen density (PSAD), transitional zone-based prostate-specific antigen density (TZ-PSAD) and lesion-based prostate-specific antigen density (lesion-PSAD) are calculated using prostate volume, transitional zone volume and lesion volume. Utilizing the aforementioned data, machine learning predictive models for clinically-significant prostate cancer (csPCa) are developed and validated.Read full original descriptionShow less
The Development and Validation of MRI-AI-based Predictive Models for csPCa
This study retrospectively included patients who underwent prostate magnetic resonance imaging (MRI) and subsequent ultrasound-guided prostate biopsy at Peking University First Hospital from January 2019 to December 2023, and prospectively enrolls patients from January 2024 to December 2029. Clinical information such as age, PSA levels, PI-RADS scores, and digital rectal examination findings are collected. A well-performing artificial intelligence model is employed to measure prostate volume, transitional zone volume, and lesion volume using MRI images. Furthermore, prostate-specific antigen density (PSAD), transitional zone-based prostate-specific antigen density (TZ-PSAD) and lesion-based prostate-specific antigen density (lesion-PSAD) are calculated using prostate volume, transitional zone volume and lesion volume. Utilizing the aforementioned data, machine learning predictive models for clinically-significant prostate cancer (csPCa) are developed and validated.
Inclusion: The interval between prostate MRI and biopsy within 3 months; Integrity of related data.
Exclusion: Any treatment for PCa prior to either MRI or biopsy, including radical prostatectomy, radiotherapy, chemotherapy, and endocrine therapy; Previous history of surgical treatment or 5α-reductase inhibitor therapy for benign prostatic hyperplasia; PSA less than 50ng/ml; Subjects undergoing MRI with an indwelling urinary catheter or suprapubic catheter; Inadequate quality of MRI images.