Universal Loss Reweighting to Balance Lesion Size Inequality in 3D
Medical Image Segmentation
- URL: http://arxiv.org/abs/2007.10033v1
- Date: Mon, 20 Jul 2020 12:08:22 GMT
- Title: Universal Loss Reweighting to Balance Lesion Size Inequality in 3D
Medical Image Segmentation
- Authors: Boris Shirokikh, Alexey Shevtsov, Anvar Kurmukov, Alexandra Dalechina,
Egor Krivov, Valery Kostjuchenko, Andrey Golanov, Mikhail Belyaev
- Abstract summary: We propose a loss reweighting approach to increase the ability of the network to detect small lesions.
We report the benefit from our method for well-known loss functions, including Dice Loss, Focal Loss, and Asymmetric Similarity Loss.
Our experiments show that inverse weighting considerably increases the detection quality, while preserves the delineation quality on a state-of-the-art level.
- Score: 50.63623720394348
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Target imbalance affects the performance of recent deep learning methods in
many medical image segmentation tasks. It is a twofold problem: class imbalance
- positive class (lesion) size compared to negative class (non-lesion) size;
lesion size imbalance - large lesions overshadows small ones (in the case of
multiple lesions per image). While the former was addressed in multiple works,
the latter lacks investigation. We propose a loss reweighting approach to
increase the ability of the network to detect small lesions. During the
learning process, we assign a weight to every image voxel. The assigned weights
are inversely proportional to the lesion volume, thus smaller lesions get
larger weights. We report the benefit from our method for well-known loss
functions, including Dice Loss, Focal Loss, and Asymmetric Similarity Loss.
Additionally, we compare our results with other reweighting techniques:
Weighted Cross-Entropy and Generalized Dice Loss. Our experiments show that
inverse weighting considerably increases the detection quality, while preserves
the delineation quality on a state-of-the-art level. We publish a complete
experimental pipeline for two publicly available datasets of CT images: LiTS
and LUNA16 (https://github.com/neuro-ml/inverse_weighting). We also show
results on a private database of MR images for the task of multiple brain
metastases delineation.
Related papers
- Brain Tumor Classification on MRI in Light of Molecular Markers [61.77272414423481]
Co-deletion of the 1p/19q gene is associated with clinical outcomes in low-grade gliomas.
This study aims to utilize a specially MRI-based convolutional neural network for brain cancer detection.
arXiv Detail & Related papers (2024-09-29T07:04:26Z) - Contour-weighted loss for class-imbalanced image segmentation [2.183832403223894]
Image segmentation is critically important in almost all medical image analysis for automatic interpretations and processing.
It is often challenging to perform image segmentation due to data imbalance between intra- and inter-class.
We propose a new methodology to address the issue, with a compact yet effective contour-weighted loss function.
arXiv Detail & Related papers (2024-06-07T07:43:52Z) - All Sizes Matter: Improving Volumetric Brain Segmentation on Small
Lesions [10.713888034128496]
We develop an ensemble of neural networks explicitly fo cused on detecting and segmenting small BMs.
We use blob loss that specifically addresses the imbalance of lesion instances in terms of size and texture and is, therefore, not biased towards larger lesions.
Our experiments demonstrate the utility of the ad ditional blob loss and the subtraction sequence.
arXiv Detail & Related papers (2023-10-04T13:56:32Z) - Robust T-Loss for Medical Image Segmentation [56.524774292536264]
This paper presents a new robust loss function, the T-Loss, for medical image segmentation.
The proposed loss is based on the negative log-likelihood of the Student-t distribution and can effectively handle outliers in the data.
Our experiments show that the T-Loss outperforms traditional loss functions in terms of dice scores on two public medical datasets.
arXiv Detail & Related papers (2023-06-01T14:49:40Z) - Random Weights Networks Work as Loss Prior Constraint for Image
Restoration [50.80507007507757]
We present our belief Random Weights Networks can be Acted as Loss Prior Constraint for Image Restoration''
Our belief can be directly inserted into existing networks without any training and testing computational cost.
To emphasize, our main focus is to spark the realms of loss function and save their current neglected status.
arXiv Detail & Related papers (2023-03-29T03:43:51Z) - On the Optimal Combination of Cross-Entropy and Soft Dice Losses for
Lesion Segmentation with Out-of-Distribution Robustness [15.08731999725517]
We study the impact of different loss functions on lesion segmentation from medical images.
We analyze the impact of the minimization of different loss functions on in-distribution performance.
Our findings are surprising: CE-Dice loss combinations that excel in segmenting in-distribution images have a poor performance when dealing with Out-of-Distribution data.
arXiv Detail & Related papers (2022-09-13T15:32:32Z) - Optimizing Operating Points for High Performance Lesion Detection and
Segmentation Using Lesion Size Reweighting [1.0514231683620514]
We propose a novel reweighing strategy to increase small pathology detection performance while maintaining segmentation accuracy.
We show that our reweighing strategy vastly outperforms competing strategies based on experiments on a large scale, multi-scanner, multi-center dataset of Multiple Sclerosis patient images.
arXiv Detail & Related papers (2021-07-27T17:43:49Z) - On the Robustness of Pretraining and Self-Supervision for a Deep
Learning-based Analysis of Diabetic Retinopathy [70.71457102672545]
We compare the impact of different training procedures for diabetic retinopathy grading.
We investigate different aspects such as quantitative performance, statistics of the learned feature representations, interpretability and robustness to image distortions.
Our results indicate that models from ImageNet pretraining report a significant increase in performance, generalization and robustness to image distortions.
arXiv Detail & Related papers (2021-06-25T08:32:45Z) - A Mixed Focal Loss Function for Handling Class Imbalanced Medical Image
Segmentation [0.7619404259039283]
We propose a new compound loss function derived from modified variants of the Focal Focal loss and Dice loss functions.
Our proposed loss function is associated with a better recall-precision balance, significantly outperforming the other loss functions in both binary and multi-class image segmentation.
arXiv Detail & Related papers (2021-02-08T20:47:38Z) - Scale-Space Autoencoders for Unsupervised Anomaly Segmentation in Brain
MRI [47.26574993639482]
We show improved anomaly segmentation performance and the general capability to obtain much more crisp reconstructions of input data at native resolution.
The modeling of the laplacian pyramid further enables the delineation and aggregation of lesions at multiple scales.
arXiv Detail & Related papers (2020-06-23T09:20:42Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.