DpDNet: An Dual-Prompt-Driven Network for Universal PET-CT Segmentation

2025-07-08

AI Summary

PET-CT lesion segmentation faces challenges including noise sensitivity, small and morphologically heterogeneous lesions, and interference from physiological high-metabolism regions; existing unified single-task approaches neglect cancer-type specificity. To address this, we propose DpDNet, a dual-prompt-driven network that jointly models cancer-specific and cross-cancer prompts to explicitly disentangle and integrate both unique and shared representations across cancer types. We introduce a prompt-aware decoding head to mitigate early-stage prompt-induced feature forgetting and adopt a dual-branch prompt architecture with end-to-end multi-task training. Evaluated on PET-CT data from four cancer types, DpDNet significantly outperforms state-of-the-art methods. Furthermore, it successfully transfers to breast cancer survival analysis, markedly improving risk stratification performance.

Technology Category

• Computer Vision: Multi-modal Vision
• Natural Language Processing: Prompt Engineering / Prompting
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

PET-CT lesion segmentation is challenging due to noise sensitivity, small and variable lesion morphology, and interference from physiological high-metabolic signals. Current mainstream approaches follow the practice of one network solving the segmentation of multiple cancer lesions by treating all cancers as a single task. However, this overlooks the unique characteristics of different cancer types. Considering the specificity and similarity of different cancers in terms of metastatic patterns, organ preferences, and FDG uptake intensity, we propose DpDNet, a Dual-Prompt-Driven network that incorporates specific prompts to capture cancer-specific features and common prompts to retain shared knowledge. Additionally, to mitigate information forgetting caused by the early introduction of prompts, prompt-aware heads are employed after the decoder to adaptively handle multiple segmentation tasks. Experiments on a PET-CT dataset with four cancer types show that DpDNet outperforms state-of-the-art models. Finally, based on the segmentation results, we calculated MTV, TLG, and SUVmax for breast cancer survival analysis. The results suggest that DpDNet has the potential to serve as a valuable tool for personalized risk stratification, supporting clinicians in optimizing treatment strategies and improving outcomes. Code is available at https://github.com/XinglongLiang08/DpDNet.

Problem

• Addresses PET-CT lesion segmentation challenges like noise and small lesions
• Overcomes limitations of treating all cancers as a single task
• Enables personalized risk stratification via improved segmentation accuracy

Innovation

• Dual-Prompt-Driven network for cancer segmentation
• Prompt-aware heads to handle multiple tasks
• Adaptive prompts for specific and shared features