QP-Prostate

QP-Prostate® is an AI-based software tool developed by Quibim for the analysis and interpretation of prostate MRI. It is designed for radiologists and urologists involved in prostate cancer detection, diagnosis, and treatment planning. The software automates PI-RADS v2.1 compliant reporting, prostate segmentation, and lesion detection, and holds FDA 510(k) clearance, CE marking, and UKCA marking.

Prostate cancer is the second most common cancer in men, and growing imaging demand has outpaced the availability of radiology expertise, leading to diagnostic delays and inconsistent application of PI-RADS v2.1 guidelines. QP-Prostate addresses this by automating key steps in the prostate MRI workflow, supporting faster and more consistent assessments across clinical settings.

Core Diagnostic Capabilities

• Prostate segmentation: Automatically segments individual prostate regions, including the peripheral zone, transitional and central zone, and seminal vesicles.
• Lesion detection and diagnosis: Identifies and highlights prostate regions suspicious for aggressive prostate cancer, assigning each a confidence-based classification score to stratify intermediate and high-grade lesions.
• Image quality assessment (AI-QUAL™): Automates prostate MRI image quality scoring based on PI-QUAL v2 guidelines, including detection of image-degrading factors such as rectal gas distortion and metal artifacts.
• Structured reporting: Generates PI-RADS compliant structured reports that automatically integrate AI findings with radiologist annotations to support clinical communication.
• Fusion biopsy planning: Integrates with most fusion biopsy devices to support more accurate targeted biopsy procedures.
• Focal therapy planning: Supports planning of focal therapy treatment through an accurate, editable 3D representation of the prostate.

AI-QUAL™ Image Quality Assessment

• Automates quality scoring based on PI-QUAL v2 guidelines, removing the need for manual image quality checks.
• Delivers standardised quality scores to support consistent reporting.
• Detects key artifacts that degrade image quality, including rectal gas distortion and metal artifacts.
• Designed to integrate into existing radiology workflows without disruption.

Workflow and User Interface

• Upgraded user interface that guarantees access to the original image at all times.
• Flexible overlays for AI-generated findings that are optional, adjustable, and removable, preserving clinician autonomy.
• Editing tools allow radiologists to modify AI outputs as needed.
• Advanced viewing with 3D visualisation of prostate anatomy and lesions.
• Integrates with PACS and fusion biopsy systems to support a fully automated end-to-end workflow.

Clinical Evidence

• A pilot study conducted in 2023 at Massachusetts General Hospital reported a 10% increase in sensitivity, with QP-Prostate increasing expert radiologists' detection rates by up to 90%.
• Algorithms are validated using a dual-validation approach incorporating both systematic and targeted biopsy data.
• Published clinical references include Jimenez-Pastor A, et al. Eur Radiol. 2023;33(7):5087–5096, and Sánchez Iglesias Á, et al. Cancers (Basel). 2023;15(16):4163.
• Clinical case examples demonstrate concordance between QP-Prostate lesion detection and targeted biopsy outcomes, including cases where the AI identified lesions missed by the radiologist and correctly classified benign tissue.

Technical and Regulatory Details

• Compatible with standard radiology workstations and supports visualisation and analysis of DICOM data.
• Regulatory clearances include FDA 510(k), CE marking, and UKCA marking.
• Developed by Quibim, with a noted partnership with Philips for AI-based MRI imaging and reporting solutions for prostate exams.