Philips Research France, 33, rue de Verdun, 92150 Suresnes, France; LTCI, Télécom ParisTech, université Paris-Saclay, 46, rue Barrault, 75013 Paris, France.
Inserm U1206, INSA-Lyon, Claude-Bernard-Lyon 1 University, CREATIS, CNRS UMR 5220, 69100 Villeurbanne, France; Department of Radiology, hospices civils de Lyon, 69002 Lyon, France.
Diagn Interv Imaging. 2019 Apr;100(4):235-242. doi: 10.1016/j.diii.2019.03.002. Epub 2019 Mar 23.
This work presents our contribution to a data challenge organized by the French Radiology Society during the Journées Francophones de Radiologie in October 2018. This challenge consisted in classifying MR images of the knee with respect to the presence of tears in the knee menisci, on meniscal tear location, and meniscal tear orientation.
We trained a mask region-based convolutional neural network (R-CNN) to explicitly localize normal and torn menisci, made it more robust with ensemble aggregation, and cascaded it into a shallow ConvNet to classify the orientation of the tear.
Our approach predicted accurately tears in the database provided for the challenge. This strategy yielded a weighted AUC score of 0.906 for all three tasks, ranking first in this challenge.
The extension of the database or the use of 3D data could contribute to further improve the performances especially for non-typical cases of extensively damaged menisci or multiple tears.
本研究旨在介绍我们在 2018 年 10 月法国放射学会举办的法国放射学年会期间参与的一项数据挑战赛的成果。该挑战赛要求根据膝关节半月板撕裂的存在情况、撕裂位置和撕裂方向对膝关节的磁共振图像进行分类。
我们训练了一个基于掩模区域的卷积神经网络（R-CNN）来明确定位正常和撕裂的半月板，并通过集成聚合使其更稳健，然后将其级联到浅层 ConvNet 中以分类撕裂的方向。
我们的方法准确地预测了挑战赛中提供的数据库中的撕裂。该策略在所有三个任务中的加权 AUC 评分为 0.906，在该挑战赛中排名第一。
扩展数据库或使用 3D 数据可能有助于进一步提高性能，特别是对于广泛受损的半月板或多个撕裂的非典型病例。
