From 1600235125d2ccc9bc2609eb7c88ecce22856dff Mon Sep 17 00:00:00 2001 From: neuronflow <7048826+neuronflow@users.noreply.github.com> Date: Wed, 27 Nov 2024 10:27:31 +0100 Subject: [PATCH] Update README.md typos --- README.md | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/README.md b/README.md index 6df070a..380f1ea 100644 --- a/README.md +++ b/README.md @@ -7,18 +7,18 @@ [![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0) -Quality prediction for brain tumor segmentation on scale ranging from 1 ⭐ star to 6 ⭐⭐⭐⭐⭐⭐ stars inspyred by the paper [**Deep Quality Estimation: Creating Surrogate Models for Human Quality Ratings**](https://arxiv.org/abs/2205.10355). -This can be used to estimate the quality of a BraTS glioma segmentation for evaluation purposes or as e.g. as part of a loss function during model training. +Quality prediction for brain tumor segmentation on a scale ranging from 1 ⭐ star to 6 ⭐⭐⭐⭐⭐⭐ stars inspired by the paper [**Deep Quality Estimation: Creating Surrogate Models for Human Quality Ratings**](https://arxiv.org/abs/2205.10355). +This can be used to estimate the quality of a BraTS glioma segmentation for evaluation purposes or as, e.g., as part of a loss function during model training. > [!NOTE] > This package expects images in atlas space and segmentation labels in brats style, i.e. > - `label 1` is the necrotic and non-enhancing tumor core > - `label 2` is the peritumoral edema -> - `label 3` is the GD-enhancing tumor (used to be `label 4` in older data, both are supported) +> - `label 3` is the GD-enhancing tumor (used to be `label 4` in older data; both are supported) > [!NOTE] -The model here is not the original model presented in the paper. Instead it is trained on the individual radiologists' ratings. This way the model has a chance to learn the variance between radioligsts' estimates. It outperforms the model presented in the paper on the test set. +The model here is not the original model presented in the paper. Instead, it is trained based on individual radiologists' ratings. This way, the model has a chance to learn the variance between radiologists' estimates. It outperforms the model presented in the paper on the test set. ## Installation