Li Hongwei Bran, Conte Gian Marco, Hu Qingqiao, Anwar Syed Muhammad, Kofler Florian, Ezhov Ivan, van Leemput Koen, Piraud Marie, Diaz Maria, Cole Byrone, Calabrese Evan, Rudie Jeff, Meissen Felix, Adewole Maruf, Janas Anastasia, Kazerooni Anahita Fathi, LaBella Dominic, Moawad Ahmed W, Farahani Keyvan, Eddy James, Bergquist Timothy, Chung Verena, Shinohara Russell Takeshi, Dako Farouk, Wiggins Walter, Reitman Zachary, Wang Chunhao, Liu Xinyang, Jiang Zhifan, Familiar Ariana, Johanson Elaine, Meier Zeke, Davatzikos Christos, Freymann John, Kirby Justin, Bilello Michel, Fathallah-Shaykh Hassan M, Wiest Roland, Kirschke Jan, Colen Rivka R, Kotrotsou Aikaterini, Lamontagne Pamela, Marcus Daniel, Milchenko Mikhail, Nazeri Arash, Weber Marc-André, Mahajan Abhishek, Mohan Suyash, Mongan John, Hess Christopher, Cha Soonmee, Villanueva-Meyer Javier, Colak Errol, Crivellaro Priscila, Jakab Andras, Albrecht Jake, Anazodo Udunna, Aboian Mariam, Yu Thomas, Chung Verena, Bergquist Timothy, Eddy James, Albrecht Jake, Baid Ujjwal, Bakas Spyridon, Linguraru Marius George, Menze Bjoern, Iglesias Juan Eugenio, Wiestler Benedikt
University of Zurich, Switzerland.
Department of Informatics, Technical University Munich, Germany.
ArXiv. 2024 Nov 24:arXiv:2305.09011v6.
Automated brain tumor segmentation methods have become well-established and reached performance levels offering clear clinical utility. These methods typically rely on four input magnetic resonance imaging (MRI) modalities: T1-weighted images with and without contrast enhancement, T2-weighted images, and FLAIR images. However, some sequences are often missing in clinical practice due to time constraints or image artifacts, such as patient motion. Consequently, the ability to substitute missing modalities and gain segmentation performance is highly desirable and necessary for the broader adoption of these algorithms in the clinical routine. In this work, we present the establishment of the Brain MR Image Synthesis Benchmark (BraSyn) in conjunction with the Medical Image Computing and Computer-Assisted Intervention (MICCAI) 2023. The primary objective of this challenge is to evaluate image synthesis methods that can realistically generate missing MRI modalities when multiple available images are provided. The ultimate aim is to facilitate automated brain tumor segmentation pipelines. The image dataset used in the benchmark is diverse and multi-modal, created through collaboration with various hospitals and research institutions.
自动化脑肿瘤分割方法已经成熟，并达到了具有明确临床实用性的性能水平。这些方法通常依赖于四种输入磁共振成像（MRI）模态：有和没有对比增强的T1加权图像、T2加权图像和FLAIR图像。然而，由于时间限制或图像伪影（如患者运动），临床实践中某些序列常常缺失。因此，对于这些算法在临床常规中的更广泛应用而言，替代缺失模态并提高分割性能的能力是非常可取且必要的。在这项工作中，我们结合医学图像计算与计算机辅助干预（MICCAI）2023会议，介绍了脑磁共振图像合成基准（BraSyn）的建立。这项挑战的主要目标是评估在提供多个可用图像时能够逼真地生成缺失MRI模态的图像合成方法。最终目的是促进自动化脑肿瘤分割流程。基准中使用的图像数据集多样且多模态，是通过与多家医院和研究机构合作创建的。
