Cancer-Net PCa-MultiSeg: Multimodal Enhancement of Prostate Cancer Lesion Segmentation Using Synthetic Correlated Diffusion Imaging

2025-11-11

AI Summary

Current prostate cancer lesion segmentation models exhibit limited performance across large-scale cohorts (Dice ≤ 0.32). To address this, we propose synthetic Correlation-based Diffusion Imaging (CDI$^s$), a plug-and-play multimodal input modality that requires no additional scan time and complements conventional diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC) maps. Leveraging co-registered multimodal images from 200 patients, we systematically evaluate six state-of-the-art segmentation architectures. Results show that CDI$^s$+DWI improves performance significantly in 50% of models without degradation; 94% of model-configurations maintain or enhance segmentation accuracy, achieving up to a 72.5% statistically significant relative Dice gain (p < 0.01). By strengthening representation robustness, CDI$^s$ overcomes inherent limitations of existing clinical diffusion modalities and delivers a deployable solution for multi-center generalization.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Security and Privacy: Data transparency and provenance
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Current deep learning approaches for prostate cancer lesion segmentation achieve limited performance, with Dice scores of 0.32 or lower in large patient cohorts. To address this limitation, we investigate synthetic correlated diffusion imaging (CDI$^s$) as an enhancement to standard diffusion-based protocols. We conduct a comprehensive evaluation across six state-of-the-art segmentation architectures using 200 patients with co-registered CDI$^s$, diffusion-weighted imaging (DWI) and apparent diffusion coefficient (ADC) sequences. We demonstrate that CDI$^s$ integration reliably enhances or preserves segmentation performance in 94% of evaluated configurations, with individual architectures achieving up to 72.5% statistically significant relative improvement over baseline modalities. CDI$^s$ + DWI emerges as the safest enhancement pathway, achieving significant improvements in half of evaluated architectures with zero instances of degradation. Since CDI$^s$ derives from existing DWI acquisitions without requiring additional scan time or architectural modifications, it enables immediate deployment in clinical workflows. Our results establish validated integration pathways for CDI$^s$ as a practical drop-in enhancement for PCa lesion segmentation tasks across diverse deep learning architectures.

Problem

• Enhancing prostate cancer lesion segmentation using synthetic correlated diffusion imaging
• Improving limited deep learning performance with CDI$^s$ multimodal enhancement
• Enabling clinical deployment without additional scan time or architectural modifications

Innovation

• Uses synthetic correlated diffusion imaging enhancement
• Integrates CDI with DWI and ADC sequences
• Derives from existing scans without extra time