• Corpus ID: 245334392

QU-BraTS: MICCAI BraTS 2020 Challenge on Quantifying Uncertainty in Brain Tumor Segmentation - Analysis of Ranking Metrics and Benchmarking Results

  title={QU-BraTS: MICCAI BraTS 2020 Challenge on Quantifying Uncertainty in Brain Tumor Segmentation - Analysis of Ranking Metrics and Benchmarking Results},
  author={Raghav Mehta and Angelos Filos and Ujjwal Baid and Chiharu Sako and Richard McKinley and Michael Rebsamen and K. Datwyler and Raphael Meier and P. Radojewski and Gowtham Krishnan Murugesan and S. Nalawade and Chandan Ganesh and Benjamin C. Wagner and Fang Frank Yu and Baowei Fei and Ananth J. Madhuranthakam and Joseph A. Maldjian and Laura Alexandra Daza and Catalina G'omez and Pablo Arbel'aez and Chengliang Dai and Shuo Wang and Hadrien Raynaud and Yuanhan Mo and Elsa D. Angelini and Yike Guo and Wenjia Bai and Subhashis Banerjee and Linmin Pei and A Murat and Sarahi Rosas-Gonz'alez and Illyess Zemmoura and Clovis Tauber and Minh H. Vu and Tufve Nyholm and Tommy Lofstedt and Laura Mora Ballestar and Ver{\'o}nica Vilaplana and Hugh McHugh and Gonzalo D. Maso Talou and Alan Wang and Jay B. Patel and Ken Chang and Katharina Hoebel and Mishka Gidwani and Nishanth Thumbavanam Arun and Sharut Gupta and Mehak Aggarwal and Praveer Singh and Elizabeth R. Gerstner and Jayashree Kalpathy-Cramer and Nicolas Boutry and Alexis Huard and Lasitha S. Vidyaratne and Md Monibor Rahman and Khan M. Iftekharuddin and Joseph Chazalon and {\'E}lodie Puybareau and Guillaume Tochon and Jun Ma and Mariano Cabezas and Xavier Llad{\'o} and Arnau Oliver and Liliana Valencia and Sergi Valverde and Mehdi Amian and Mohammadreza Soltaninejad and Andriy Myronenko and Ali Hatamizadeh and Xuejing Feng and Quan Dou and N. Tustison and Craig Meyer and Nisarg A. Shah and Sanjay N. Talbar and M. A. Weber and Abhishek Mahajan and Andr{\'a}s Jakab and Roland Wiest and Hassan M. Fathallah-Shaykh and Arash Nazeri and Mikhail Milchenko and Daniel Marcus and Aikaterini Kotrotsou and Rivka R. Colen and John B. Freymann and Justin S. Kirby and Christos Davatzikos and Bjoern H Menze and Spyridon Bakas and Yarin Gal and Tal Arbel},
Deep learning (DL) models have provided state-of-the-art performance in various medical imaging benchmarking challenges, including the Brain Tumor Segmentation (BraTS) challenges. However, the task of focal pathology multi-compartment segmentation (e.g., tumor and lesion sub-regions) is particularly challenging, and potential errors hinder translating DL models into clinical workflows. Quantifying the reliability of DL model predictions in the form of uncertainties could enable clinical review… 

Trustworthy clinical AI solutions: a unified review of uncertainty quantification in deep learning models for medical image analysis

An overview of the existing methods to quantify uncertainty associated to DL predictions is proposed, which focuses on applications to medical image analysis, which present specific challenges due to the high dimensionality of images and their quality variability, as well as constraints associated to real-life clinical routine.

Novel structural-scale uncertainty measures and error retention curves: application to multiple sclerosis

This paper focuses on the uncertainty estimation for white matter lesions (WML) segmentation in magnetic resonance imaging (MRI). On one side, voxel-scale segmentation errors cause the erroneous

Region-Based Evidential Deep Learning to Quantify Uncertainty and Improve Robustness of Brain Tumor Segmentation

This paper proposes a region-based EDL segmentation framework that can generate reliable uncertainty maps and accurate segmentation results, which is robust to noise and image corruption and showed the potential for clinical application.

Quantifying U-Net Uncertainty in Multi-Parametric MRI-based Glioma Segmentation by Spherical Image Projection

To develop a U-Net segmentation uncertainty quantification method based on spherical image projection of multi-parametric MRI (MP-MRI) in glioma segmentation. Methods: The projection of planar MRI



MRI brain tumor segmentation and uncertainty estimation using 3D-UNet architectures

3D encoder-decoder architectures trained with patchbased techniques to reduce memory consumption and decrease the effect of unbalanced data are studied and voxel-wise uncertainty information is introduced, both epistemic and aleatoric using test-time dropout and data-augmentation respectively.

The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS)

The set-up and results of the Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) organized in conjunction with the MICCAI 2012 and 2013 conferences are reported, finding that different algorithms worked best for different sub-regions, but that no single algorithm ranked in the top for all sub-Regions simultaneously.

Exploring Uncertainty Measures in Deep Networks for Multiple Sclerosis Lesion Detection and Segmentation

A 3D MS lesion segmentation CNN is developed, augmented to provide four different voxel-based uncertainty measures based on MC dropout, and the first exploration of multiple uncertainty estimates based on Monte Carlo (MC) dropout is presented in the context of deep networks for lesion detection and segmentation in medical images.

Analyzing the Quality and Challenges of Uncertainty Estimations for Brain Tumor Segmentation

An analysis of the most commonly used uncertainty estimation methods in regards to benefits and challenges for brain tumor segmentation, and shows that the uncertainty methods are typically well-calibrated when evaluated at the dataset level.

Brain Tumor Segmentation with Uncertainty Estimation and Overall Survival Prediction

A deep-learning-based segmentation method using an ensemble of 3D U-Nets with different hyper-parameters, which can effectively reduce overfitting and regression errors and a linear model for survival prediction using extracted imaging and non-imaging features are developed.

Uncertainty Evaluation Metric for Brain Tumour Segmentation

A metric to assess and rank uncertainty measures for the task of brain tumour sub-tissue segmentation in the BraTS 2019 sub-challenge on uncertainty quantification is developed and the workings of the components of the metric are explored.

ISLES 2015 ‐ A public evaluation benchmark for ischemic stroke lesion segmentation from multispectral MRI

Multi-Decoder Networks with Multi-Denoising Inputs for Tumor Segmentation

An end-to-end deep-learning-based segmentation method using a multi-decoder architecture by jointly learning three separate sub-problems using a partly shared encoder and applying smoothing methods to the input images to generate denoised versions as additional inputs to the network is developed.

3D U-Net for Brain Tumour Segmentation

A modified version of the popular 3D U-net architecture, which takes as input multi-modal brain MR volumes, processes them at multiple scales, and generates a full resolution multi-class tumour segmentation as output.