Evaluation of MRI to Ultrasound Registration Methods for Brain Shift Correction: The CuRIOUS2018 Challenge

Abstract

In brain tumor surgery, the quality and safety of the procedure can be impacted by intra-operative tissue deformation, called brain shift. Brain shift can move the surgical targets and other vital structures such as blood vessels, thus invalidating the presurgical plan. Intra-operative ultrasound (iUS) is a convenient and cost-effective imaging tool to track brain shift and tumor resection. Accurate image registration techniques that update pre-surgical MRI based on iUS are crucial but challenging. The MICCAI Challenge 2018 for Correction of Brain shift with Intra-Operative UltraSound (CuRIOUS2018) provided a public platform to benchmark MRI-iUS registration algorithms on newly released clinical datasets. In this work, we present the data, setup, evaluation, and results of CuRIOUS 2018, which received 6 fully automated algorithms from leading academic and industrial research groups. All algorithms were first trained with the public RESECT database, and then ranked based on a test dataset of 10 additional cases with identical data curation and annotation protocols as the RESECT database. The article compares the results of all participating teams and discusses the insights gained from the challenge, as well as future work.

Deformable MRI-Ultrasound registration using correlation-based attribute matching for brain shift correction: Accuracy and generality in multi-site data

“…Besides efforts in algorithm development aiming for accuracy and generality, the evaluation of sixteen algorithms in three datasets is the most comprehensive so far in the MR-US registration tasks. CuRIOUS in 2018 marks so-far the first independent evaluation of MR-US registration algorithms in a well-organized open platform that consists of training and testing data (Xiao et al, 2019). Our work uses results from the CuRIOUS 2018 Challenge (Table 4, results for the RESECT dataset).…”

“…Our work uses results from the CuRIOUS 2018 Challenge (Table 4, results for the RESECT dataset). The extensions are: (a) we used two additional datasets to make it multi-site evaluation (details in Table 1); therefore, not only the accuracy but also the generality of sixteen algorithms (Table 2, compared to six algorithms included in (Xiao et al, 2019), some with multiple sets of parameter configurations (Table 3), can be further evaluated in multi-site datasets (Table 4 for accuracy and generality); and (b) we further annotated the regions of the landmark pairs, which led to the first report of the registration accuracy by the extent of brain shift that can be corrected ( Figure 10), and the registration accuracy by brain regions (Figures 11 and 12) and by tumor grades (Figures 13).…”

“…The Correction of Brain shift with Intra-Operative Ultrasound (CuRIOUS) 2018 Challenge (http://curious2018.grand-challenge.org) was held in conjunction with the Medical Image Computing and Computer Assisted Intervention (MICCAI) conference (Xiao et al, 2019). This Challenge included two phases.…”

Correlated Multimodal Imaging in Life Sciences: Expanding the Biomedical Horizon

“…Challenges are the competition of algorithms on a given dataset, where the ground truth is known in order to rank the algorithms according to a set of metrics designed for the challenge. Some challenges have been organized for the multimodal field, but usually focused on one particular problem (Anhir focused on multiplexed histological data or Curious for US to MRI brain images registration [145]). These challenges are of great interest since they provide a way to identify the actual level of accuracy state of the art and give direction for future research.…”

The DIRECT algorithm: 25 years Later

“…DIRECT has also proven effective on low-dimensional, real-world problems from a wide range of fields, including genetics [42], surgery [76], air transport [8], power generation [2,13,34], astronomy [4], hydrogen fuel cell management [74], photovoltaic systems [49], hard disk design [77], aircraft routing [1], and many others [3,6,20,24,35,44,58,71]. However, DIRECT has been found to be less competitive on higher-dimensional problems [47].…”