基于解剖结构的AI系统可自动评估前列腺癌风险并解释决策依据。
Explainable Anatomy-Guided AI for Prostate MRI: Foundation Models and In Silico Clinical Trials for Virtual Biopsy-based Risk Assessment
- 用解剖先验指导深度学习，结合3D影像与临床数据进行癌症风险分层。
- 在617例数据上达到AUC 0.79，诊断准确率提升至0.77，时间节省40%。
- 生成反事实热图揭示关键病变区域，适合临床医生辅助诊断使用。
我们提出一种全自动、解剖引导的深度学习流程，用于基于常规MRI进行前列腺癌风险分层。该流程包含三个核心组件：使用nnU-Net分割前列腺及其各分区（轴向T2加权影像）；基于UMedPT Swin Transformer基础模型的分类模块，利用3D图像块并可选加入解剖先验和临床数据微调；以及基于VAE-GAN的反事实热图生成框架，定位驱动决策的图像区域。系统训练使用1,500例PI-CAI病例进行分割，617例双参数MRI及来自CHAIMELEON挑战赛的元数据用于分类（70%训练，10%验证，20%测试）。分割平均Dice分数分别为0.95（腺体）、0.94（外周带）、0.92（移行带）。引入腺体先验后AUC从0.69升至0.72，三尺度集成达最优性能（AUC=0.79，综合评分=0.76），优于2024年CHAIMELEON挑战赛冠军。反事实热图可靠地突出分割区域内病灶，提升模型可解释性。在前瞻性多中心模拟临床试验中，20名医生使用AI后诊断准确率由0.72提升至0.77，Cohen's kappa从0.43增至0.53，单例审查时间减少40%。结果表明，具备反事实解释能力的解剖感知基础模型可实现精准、可解释且高效的前列腺癌风险评估，具备作为虚拟活检应用于临床的潜力。
原文摘要 · Abstract (English)
We present a fully automated, anatomically guided deep learning pipeline for prostate cancer (PCa) risk stratification using routine MRI. The pipeline integrates three key components: an nnU-Net module for segmenting the prostate gland and its zones on axial T2-weighted MRI; a classification module based on the UMedPT Swin Transformer foundation model, fine-tuned on 3D patches with optional anatomical priors and clinical data; and a VAE-GAN framework for generating counterfactual heatmaps that localize decision-driving image regions. The system was developed using 1,500 PI-CAI cases for segmentation and 617 biparametric MRIs with metadata from the CHAIMELEON challenge for classification (split into 70% training, 10% validation, and 20% testing). Segmentation achieved mean Dice scores of 0.95 (gland), 0.94 (peripheral zone), and 0.92 (transition zone). Incorporating gland priors improved AUC from 0.69 to 0.72, with a three-scale ensemble achieving top performance (AUC = 0.79, composite score = 0.76), outperforming the 2024 CHAIMELEON challenge winners. Counterfactual heatmaps reliably highlighted lesions within segmented regions, enhancing model interpretability. In a prospective multi-center in-silico trial with 20 clinicians, AI assistance increased diagnostic accuracy from 0.72 to 0.77 and Cohen's kappa from 0.43 to 0.53, while reducing review time per case by 40%. These results demonstrate that anatomy-aware foundation models with counterfactual explainability can enable accurate, interpretable, and efficient PCa risk assessment, supporting their potential use as virtual biopsies in clinical practice.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。