用语言模型引导扩散生成，从基线脑影像预测早期痴呆转化
Diffusion with a Linguistic Compass: Steering the Generation of Clinically Plausible Future sMRI Representations for Early MCI Conversion Prediction
- 用多任务去噪网络学习不规则随访数据的潜在变化轨迹
- 引入语言模型评分机制，确保生成结果符合临床病理特征
- 在两个大型队列中提升早期转化预测准确率5-12%，适合临床风险评估
早期预测轻度认知障碍（MCI）向痴呆转化面临及时性与准确性之间的权衡：单次基线结构磁共振成像（sMRI）可实现快速评估，但缺乏疾病进展信息；而纵向扫描虽能捕捉变化，却延迟诊断。本文提出MCI-Diff，一种基于扩散模型的框架，直接从基线sMRI合成具有临床合理性的未来脑影像表征，兼顾实时风险评估与高预测性能。首先，采用多任务序列重建策略，在插值与外推任务上训练共享去噪网络，以应对不规则随访采样并学习鲁棒的潜在轨迹。其次，引入大语言模型驱动的“语言罗盘”实现临床合理性采样：生成的特征候选经量化、分词后，由微调的语言模型根据预期结构生物标志物进行评分，引导自回归生成趋向真实疾病模式。在ADNI和AIBL队列上的实验表明，MCI-Diff优于现有最先进方法，早期转化预测准确率提升5%-12%。
原文摘要 · Abstract (English)
Early prediction of Mild Cognitive Impairment (MCI) conversion is hampered by a trade-off between immediacy--making fast predictions from a single baseline sMRI--and accuracy--leveraging longitudinal scans to capture disease progression. We propose MCI-Diff, a diffusion-based framework that synthesizes clinically plausible future sMRI representations directly from baseline data, achieving both real-time risk assessment and high predictive performance. First, a multi-task sequence reconstruction strategy trains a shared denoising network on interpolation and extrapolation tasks to handle irregular follow-up sampling and learn robust latent trajectories. Second, an LLM-driven "linguistic compass" is introduced for clinical plausibility sampling: generated feature candidates are quantized, tokenized, and scored by a fine-tuned language model conditioned on expected structural biomarkers, guiding autoregressive generation toward realistic disease patterns. Experiments on ADNI and AIBL cohorts show that MCI-Diff outperforms state-of-the-art baselines, improving early conversion accuracy by 5-12%.
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。