V Couteaux1, S Si-Mohamed2, O Nempont3, T Lefevre3, A Popoff3, G Pizaine3, N Villain3, I Bloch4, A Cotten5, L Boussel2. 1. Philips Research France, 33, rue de Verdun, 92150 Suresnes, France; LTCI, Télécom ParisTech, université Paris-Saclay, 46, rue Barrault, 75013 Paris, France. Electronic address: vincent.couteaux@telecom-paristech.fr. 2. 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. 3. Philips Research France, 33, rue de Verdun, 92150 Suresnes, France. 4. LTCI, Télécom ParisTech, université Paris-Saclay, 46, rue Barrault, 75013 Paris, France. 5. Department of Musculoskeletal Radiology, CHRU de Lille, 59000 Lille, France.
Abstract
PURPOSE: 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. MATERIALS AND METHODS: 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. RESULTS: 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. CONCLUSION: 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.
PURPOSE: 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. MATERIALS AND METHODS: 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. RESULTS: 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. CONCLUSION: 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.
Authors: Leah A Groves; Blake VanBerlo; Natan Veinberg; Abdulrahman Alboog; Terry M Peters; Elvis C S Chen Journal: Int J Comput Assist Radiol Surg Date: 2020-08-24 Impact factor: 2.924
Authors: Gabby B Joseph; Charles E McCulloch; Jae Ho Sohn; Valentina Pedoia; Sharmila Majumdar; Thomas M Link Journal: Skeletal Radiol Date: 2021-11-04 Impact factor: 2.199