arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2112.08845cs.LG
收藏

基于多实例学习的脑肿瘤磁共振波谱检测

Multiple Instance Learning for Brain Tumor Detection from Magnetic Resonance Spectroscopy Data

Frankfurt Institute for Advanced Studies(法兰克福高等研究院);Institute for Neuroradiology at Frankfurt university hospital(法兰克福大学医院神经放射学研究所)

Diyuan Lu, Gerhard Kurz, Nenad Polomac, Iskra Gacheva, Elke Hattingen, Jochen Triesch

摘要:

我们将深度学习(DL)应用于磁共振波谱(MRS)数据,以完成脑肿瘤检测任务。医疗应用程序经常受到数据稀缺和噪声破坏的影响。这两个问题在我们的数据集中都很突出。此外,不同数量的光谱可用于不同的患者。我们通过将任务视为多实例学习(MIL)问题来解决这些问题。具体而言,我们将来自同一患者的多个光谱聚集到一个"袋子"中进行分类,并应用数据增强技术。为了在装袋过程中实现置换不变性,我们提出了两种方法:(1)对一个袋子中所有样本的特征应用最小、最大和平均池化;(2)应用注意力机制。我们在多个神经网络架构上测试了这两种方法。我们证明了在多个实例而不是单个光谱上训练时,分类性能得到了显着提高。我们提出了一个简单的过采样数据增强方法,并表明它可以进一步提高性能。最后,我们证明了我们提出的模型优于人工分类神经放射学家根据大多数性能指标。

英文摘要:

We apply deep learning (DL) on Magnetic resonance spectroscopy (MRS) data for the task of brain tumor detection. Medical applications often suffer from data scarcity and corruption by noise. Both of these problems are prominent in our data set. Furthermore, a varying number of spectra are available for the different patients. We address these issues by considering the task as a multiple instance learning (MIL) problem. Specifically, we aggregate multiple spectra from the same patient into a "bag" for classification and apply data augmentation techniques. To achieve the permutation invariance during the process of bagging, we proposed two approaches: (1) to apply min-, max-, and average-pooling on the features of all samples in one bag and (2) to apply an attention mechanism. We tested these two approaches on multiple neural network architectures. We demonstrate that classification performance is significantly improved when training on multiple instances rather than single spectra. We propose a simple oversampling data augmentation method and show that it could further improve the performance. Finally, we demonstrate that our proposed model outperforms manual classification by neuroradiologists according to most performance metrics.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑