Prostate Cancer
Conditions
Brief summary
It is a prospective paired-cohort study for diagnostic test evaluation. The study aim to determine the accuracy of AI review and investigate whether AI review could detect MRI visible significant cancer as effective as radiologist review. MRI image of about 368 men recommended for biopsy will be reviewed by an AI model and an experienced radiologist, respectively. AI review (index) and radiologist review (standard) will be blinded to each other, while biopsy urologists will be well-informed of the findings of both AI review and radiologist review and make personalized biopsy plan by combining both findings. The pathological results of MRI-ultrasound fusion biopsy (reference) will serve as the gold standard to assess the diagnostic accuracy.
Detailed description
ProstateAI is a prospective paired-cohort study for diagnostic test evaluation. ProstateAI will determine the accuracy of AI review and investigate whether AI review could detect MRI visible significant cancer as effective as radiologist review. MRI image of about 368 men recommended for biopsy will be reviewed by an AI model and an experienced radiologist, respectively. AI review (index) and radiologist review (standard) will be blinded to each other, while biopsy urologists will be well-informed of the findings of both AI review and radiologist review and make personalized biopsy plan by combining both findings. The pathological results of MRI-ultrasound fusion biopsy (reference) will serve as the gold standard to assess the diagnostic accuracy.
Interventions
DIAGNOSTIC_TESTAI modal
The MRI image will be reviewed by radiologist and AI model(s) respectively. The urologist will combine the results of the two approaches to optimize the biopsy strategy which is expected to result in more accurate diagnosis.
DIAGNOSTIC_TESTStandard review
The MRI image will be reviewed by radiologist.
Sponsors
Chinese University of Hong Kong
Study design
Allocation
NON_RANDOMIZED
Intervention model
CROSSOVER
Primary purpose
DIAGNOSTIC
Masking
SINGLE (Caregiver)
Masking description
AI review (index) and radiologist review (standard) will be blinded to each other, while biopsy urologists will be well-informed of the findings of both AI review and radiologist review and make personalized biopsy plan by combining both findings.
Eligibility
Sex/Gender
MALE
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Patients recommended for prostate biopsy for diagnosis or re-assessment. * Men at least 18 years or over * Patients with prostate MRI image eligible for radiologist review and AI review. * Patient Informed Consent is signed.
Exclusion criteria
* Patient contraindicated to prostate biopsy * Patient failed to complete the biopsy procedure
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Prostate Cancer detection rate of ISUP grade group ≥2 | Through study completion, an average of 1 year | Assessed by AI review, Radiologist review, and Biopsy (Targeted). |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Diagnosis of any grade of prostate cancer | Through study completion, an average of 1 year | Assessed by prostate biopsy |
Countries
Hong Kong
Contacts
Primary ContactPeter Ka-Fung CHIU, FRCS, PhD
Outcome results
None listed