Negative controls reveal volume-driven confounding in radiomics and imaging foundation model features

AI 中文解读

AI医学影像分析最近爆出个“翻车”隐患：原本以为能看透肿瘤本质的AI，很可能只是在“看个头”而已。一项新研究开发了一套叫READII-2-ROQC的工具，专门给AI“打假”——它会把医学图像里的肿瘤区域或者背景像素随机打乱，但保持体积不变，然后看AI的诊断结果会不会跟着变。结果发现，很多之前声称准确的癌症预测模型，在图像被彻底打乱后依然“稳如泰山”，说明它们根本没学习到肿瘤的真实结构，只是在靠体积或扫描背景“作弊”。这一发现对普通患者意义重大：未来AI辅助诊断会更加规范，医生能分辨出哪些AI是真的在看病理特征，哪些只是按体积大小猜结果。有了这套质检工具，影像AI将不再“黑箱”操作，而是能通过严格考验的可靠助手，减少误诊风险，让精准医疗真正落到实处。

Radiomics and imaging foundation models promise non-invasive biomarkers of tumour biology, yet predictive signatures may reflect tumour volume or acquisition artifacts rather than meaningful image structure. We introduce READII-2-ROQC, an open-source framework that uses volume-preserving negative controls to assess whether radiomic and deep imaging features capture independent spatial signals. READII-2-ROQC generates voxel-perturbed images across tumour, background and whole-image regions using configurable randomization strategies, then compares feature behaviour and model performance between original and control images. Applied to three public cancer imaging cohorts, the framework processed 3,552 tumour volumes and extracted PyRadiomics and foundation-model features from original images and nine matched controls. Reproducing published survival and HPV-status signatures, we show that multiple models retain performance after spatial structure is destroyed, revealing volume-driven or contextual confounding, whereas others show perturbation-sensitive signal. READII-2-ROQC provides a scalable quality-control strategy for developing interpretable, biologically grounded imaging biomarkers and reproducible radiomics workflows.