Advances in data-driven deep learning for chest X-ray image analysis underscore the need for explainability, privacy, large datasets and significant computational resources. We frame privacy and explainability as a lossy single-image compression problem to reduce both computational and data requirements without training. For Cardiomegaly detection in chest X-ray images, we propose HeartSpot and four spatial bias priors. HeartSpot priors define how to sample pixels based on domain knowledge from medical literature and from machines. HeartSpot privatizes chest X-ray images by discarding up to 97% of pixels, such as those that reveal the shape of the thoracic cage, bones, small lesions and other sensitive features. HeartSpot priors are ante-hoc explainable and give a human-interpretable image of the preserved spatial features that clearly outlines the heart. HeartSpot offers strong compression, with up to 32x fewer pixels and 11x smaller filesize. Cardiomegaly detectors using HeartSpot are up to 9x faster to train or at least as accurate (up to +.01 AUC ROC) when compared to a baseline DenseNet121. HeartSpot is post-hoc explainable by re-using existing attribution methods without requiring access to the original non-privatized image. In summary, HeartSpot improves speed and accuracy, reduces image size, improves privacy and ensures explainability. Source code: https://www.github.com/adgaudio/HeartSpot
翻译:用于胸前X射线图像分析的数据驱动深层学习的进展突出表明,需要解释性、隐私、大型数据集和大量计算资源。我们把隐私和解释性定义为一个丢失的单一图像压缩问题,以降低计算和数据要求。对于胸前X射线图像的心血管探测,我们提议在胸后X射线图像中进行心脏波和四个空间偏差前科。心脏Spot前科根据医学文献和机器的域知识,定义如何抽样像素。心脏Spot将胸前X光图像私有化,丢弃多达97%的像素,例如那些显示胸部笼、骨骼、小损伤和其他敏感特性形状的像素。心Spot前科可以解释并给出清晰的保存空间特征的人类间图像。心脏Spot前科提供强力压缩,最多减少32个像素和11个更小的档案化。使用心脏ShesterSpoint的心血管智能探测器可以更快地或至少以准确的方式进行训练(最高为直观的直径、骨、骨、小伤、直等) 直径SlotSlot-rudi-libal 和直径Scar-liverScar-cal-cal-laft-laft-sc-laft-laft-sc-sc-laft-sc-laft-laft-sc-sc-sc-sc-sc-sc-sc-sc-lability-sc-sc-sc-sc-sc-sc-sc-lability-lability-sc-labilable-lability-to-labilable-to-to-to-to-sc-sc-lais-lais-to-to-sc-sc-sc-sc-sc-to-sc-sc-sc-sc-sc-sc-sc-sc-sc-sc-sc-sc-sc-sc-sc-sc-sc-labil-lab-lab-sc-labal-sc-sc-sc-sc-sc-lab-sc-sc-sc-sc-sc-sc-sc-sc-sc-sc-sc-sc-sc-sc-sc-sc-sc-labal-labil-lais-laut-laut-sc-la