# AI Predicts Prostate Cancer Spread from Medical Scans

> This patent describes how artificial intelligence can analyze medical scans of a prostate tumor to automatically identify suspect regions and predict if the cancer will spread to other parts of the body.

- **Patent:** US 12737884
- **Original title:** Systems and methods for AI-assisted analysis of primary tumor images for prediction of metastases
- **Granted:** 2026
- **Status:** Active
- **Times cited:** 0
- **Field:** biotech, ai_ml, software, medical_devices

## What it does

This patent details systems and methods that use artificial intelligence, specifically a convolutional neural network (CNN), to analyze medical images of the prostate. The system automatically identifies 'suspect regions' or 'hotspots' within images like PET, CT, or PET/CT scans. It then uses these identified regions to predict whether localized prostate cancer has spread or will spread, a process known as metastasis. For example, a doctor could upload a patient's prostate PET/CT scan, and the AI system would highlight areas of concern and provide a prediction on the likelihood of the cancer spreading.

## What it does NOT cover

- Does not cover predicting cancer spread from images of organs other than the prostate.
- Does not cover using different types of AI models beyond convolutional neural networks for this specific prediction.
- Does not cover predicting other cancer outcomes, such as tumor growth rate or response to treatment, only metastasis.
- Does not cover using other imaging modalities like MRI or ultrasound for this specific prediction of prostate cancer metastasis.
- Does not cover manual identification of suspect regions without AI assistance.

## The clever bit

The novel aspect is the automated identification of specific 'suspect regions' within prostate images by a convolutional neural network, which is then directly used to predict the likelihood of cancer spreading, rather than just detecting the presence of a tumor.

## Real-world examples

1. AI-powered diagnostic tools for prostate cancer staging
2. Clinical decision support systems for oncology
3. Medical imaging analysis software for predictive analytics

## Why it matters

Accurate and early prediction of cancer metastasis is critical for effective treatment planning and improving patient outcomes. This technology aims to provide clinicians with a powerful tool to assess the risk of prostate cancer spreading, potentially enabling more personalized and timely interventions. It represents a significant step towards integrating advanced AI capabilities into routine oncology diagnostics, enhancing the precision of medical prognoses.

## Frequently asked questions

### What does AI Predicts Prostate Cancer Spread from Medical Scans cover?

This patent describes how artificial intelligence can analyze medical scans of a prostate tumor to automatically identify suspect regions and predict if the cancer will spread to other parts of the body.

### When does this patent expire?

This patent is expected to expire on September 15, 2046, when the invention enters the public domain.

### What problem does this patent solve?

Accurate and early prediction of cancer metastasis is critical for effective treatment planning and improving patient outcomes. This technology aims to provide clinicians with a powerful tool to assess the risk of prostate cancer spreading, potentially enabling more personalized and timely interventions. It represents a significant step towards integrating advanced AI capabilities into routine oncology diagnostics, enhancing the precision of medical prognoses.

### What does this patent NOT cover?

Does not cover predicting cancer spread from images of organs other than the prostate.

**Full plain-English explainer:** https://patentbrief.org/patent/us/12737884/systems-and-methods-for-ai-assisted-analysis-of-primary-tumor-images-for

**Original patent:** https://patents.google.com/patent/US12737884

---

_Source: PatentBrief — https://patentbrief.org. Patent facts are from public records; the plain-English explanation is PatentBrief's._