Automated Prostate Cancer Staging via 3D PSMA-PET Analysis

Solution Overview

Problem

Current methods for determining a prostate cancer staging score from PSMA-PET images are not fully automated and are subject to variation, necessitating more reproducible and standardized reporting to support wider integration of PSMA-PET into clinical guidelines.

Innovation Solution

The use of a machine learning model, such as convolutional neural networks (CNNs), to analyze 3D images from both functional and anatomical imaging modalities, allowing for the automated determination of prostate cancer staging scores by identifying regions of PSMA binding agent uptake and associating them with specific prostate zones.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual methods are used to determine prostate cancer staging scores from PSMA-PET images, then flexibility in interpretation is maintained, but reproducibility and standardization deteriorate

Engineering Contradiction:

Improvereproducibility of staging score determinationVSAvoidcomplexity of automated system

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent replaces manual visual assessment and interpretation of PSMA-PET images with an automated machine learning system. The ML model processes imaging data to generate staging scores, substituting the mechanical/manual process of radiologist assessment with an automated computational system that ensures consistent application of scoring criteria across different patients and time points.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service by automatically performing the complete staging score determination process without requiring manual intervention. The ML model independently processes imaging data, identifies uptake regions, and generates staging scores, making the process self-sufficient while maintaining reproducibility and standardization.

Inventive Principle:

Principle #25Self-service

2Measurement precision

If automated machine learning methods are used to determine prostate cancer staging scores, then reproducibility and standardization improve, but system complexity increases

Engineering Contradiction:

Improveconsistency of staging score determinationVSAvoidcomplexity of machine learning system

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces manual visual assessment and interpretation of PSMA-PET images with an automated machine learning system. The ML model processes imaging data to generate staging scores, substituting the mechanical/manual process of radiologist assessment with an automated computational system that ensures consistent application of scoring criteria across different patients and time points.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system transforms complex imaging data into standardized numerical staging scores through the ML model. By changing the parameters from raw image data to structured scoring outputs, the system achieves measurement precision and consistency while managing complexity through parameter transformation and standardization.

Inventive Principle:

Principle #35Parameter changes

3Device complexity

If manual assessment of PSMA-PET images is used, then system simplicity is maintained, but measurement precision and reproducibility deteriorate

Engineering Contradiction:

Improvesimplicity of current manual systemVSAvoidreproducibility of staging scores

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent replaces manual visual assessment and interpretation of PSMA-PET images with an automated machine learning system. The ML model processes imaging data to generate staging scores, substituting the mechanical/manual process of radiologist assessment with an automated computational system that ensures consistent application of scoring criteria across different patients and time points.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service by automatically performing the complete staging score determination process without requiring manual intervention. The ML model independently processes imaging data, identifies uptake regions, and generates staging scores, making the process self-sufficient while maintaining reproducibility and standardization.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS20250191752A1Systems and methods for automated determination of a prostate cancer staging score

Publication Date: 2025.06.12 EXINI DIAGNOSTICS

AI summary

Presented herein are systems and methods for the automated determination of a prostate cancer staging score (e.g., a PRIMARY score) for a subject. In certain embodiments, the systems and methods employ a machine learning model (e.g., one or more convolutional neural networks, CNNs) to analyze three-dimensional (3D) images obtained via both a functional imaging modality and an anatomical imaging modality. In addition to identifying regions of PSMA binding agent uptake (hotspots), the techniques described herein are able to accurately and automatically associate specific prostate zones to each hotspot and use this information in the determination of the staging score.