Schmidt-Mengin Marius, Soulier Théodore, Hamzaoui Mariem, Yazdan-Panah Arya, Bodini Benedetta, Ayache Nicholas, Stankoff Bruno, Colliot Olivier
Institut du Cerveau-Paris Brain Institute, Centre National de la Recherche Scientifique, Inria, Inserm, Assistance Publique-Hôpitaux de Paris, Hôpital de la Pitié Salpêtrière, Sorbonne Université, Paris, France.
Institut du Cerveau-Paris Brain Institute, Centre National de la Recherche Scientifique, Inserm, Assistance Publique-Hôpitaux de Paris, Hôpital de la Pitié Salpêtrière, Sorbonne Université, Paris, France.
Front Neurosci. 2022 Nov 4;16:1004050. doi: 10.3389/fnins.2022.1004050. eCollection 2022.
Detecting new lesions is a key aspect of the radiological follow-up of patients with Multiple Sclerosis (MS), leading to eventual changes in their therapeutics. This paper presents our contribution to the MSSEG-2 MICCAI 2021 challenge. The challenge is focused on the segmentation of new MS lesions using two consecutive Fluid Attenuated Inversion Recovery (FLAIR) Magnetic Resonance Imaging (MRI). In other words, considering longitudinal data composed of two time points as input, the aim is to segment the lesional areas, which are present only in the follow-up scan and not in the baseline. The backbone of our segmentation method is a 3D UNet applied patch-wise to the images, and in which, to take into account both time points, we simply concatenate the baseline and follow-up images along the channel axis before passing them to the 3D UNet. Our key methodological contribution is the use of online hard example mining to address the challenge of class imbalance. Indeed, there are very few voxels belonging to new lesions which makes training deep-learning models difficult. Instead of using handcrafted priors like brain masks or multi-stage methods, we experiment with a novel modification to online hard example mining (OHEM), where we use an exponential moving average (i.e., its weights are updated with momentum) of the 3D UNet to mine hard examples. Using a moving average instead of the raw model should allow smoothing of its predictions and allow it to give more consistent feedback for OHEM.
检测新病灶是多发性硬化症（MS）患者放射学随访的关键环节，这最终会导致其治疗方案的改变。本文介绍了我们对MSSEG - 2 MICCAI 2021挑战赛的贡献。该挑战赛专注于使用连续两次的液体衰减反转恢复（FLAIR）磁共振成像（MRI）对新的MS病灶进行分割。换句话说，将由两个时间点组成的纵向数据作为输入，目标是分割仅在随访扫描中出现而在基线扫描中未出现的病灶区域。我们分割方法的核心是一个逐块应用于图像的3D UNet，并且为了考虑两个时间点，我们在将基线图像和随访图像传递给3D UNet之前，简单地沿通道轴将它们连接起来。我们关键的方法学贡献是使用在线困难样本挖掘来应对类别不平衡的挑战。实际上，属于新病灶的体素非常少，这使得训练深度学习模型变得困难。我们没有使用诸如脑掩码之类的手工先验信息或多阶段方法，而是对在线困难样本挖掘（OHEM）进行了一种新颖的改进实验，即我们使用3D UNet的指数移动平均值（即其权重通过动量更新）来挖掘困难样本。使用移动平均值而不是原始模型应该可以平滑其预测，并使其能够为OHEM提供更一致的反馈。
