Combo Loss: Handling Input and Output Imbalance in Multi-Organ Segmentation

By Saeid Asgari Taghanaki, Yefeng Zheng, Shaohua Kevin Zhou, Bogdan Georgescu, Puneet S. Sharma, Daguang Xu, Dorin Comaniciu, Ghassan Hamarneh. Comput. Medical Imaging Graph., 2018

434 citations recorded by the source corpus.

Simultaneous segmentation of multiple organs from different medical imaging modalities is a crucial task as it can be utilized for computer-aided diagnosis, computer-assisted surgery, and therapy planning. Thanks to the recent advances in deep learning, several deep neural networks for medical image segmentation have been introduced successfully for this purpose. In this paper, we focus on learning a deep multi-organ segmentation network that labels voxels. In particular, we examine the critical choice of a loss function in order to handle the notorious imbalance problem that plagues both the input and output of a learning model. The input imbalance refers to the class-imbalance in the input training samples (i.e., small foreground objects embedded in an abundance of background voxels, as well as organs of varying sizes). The output imbalance refers to the imbalance between the false positives and false negatives of the inference model. In order to tackle both types of imbalance during training and inference, we introduce a new curriculum learning based loss function. Specifically, we leverage Dice similarity coefficient to deter model parameters from being held at bad local minima and at the same time gradually learn better model parameters by penalizing for false positives/negatives using a cross entropy term. We evaluated the proposed loss function on three datasets: whole body positron emission tomography (PET) scans with 5 target organs, magnetic resonance imaging (MRI) prostate scans, and ultrasound echocardigraphy images with a single target organ i.e., left ventricular. We show that a simple network architecture with the proposed integrative loss function can outperform state-of-the-art methods and results of the competing methods can be improved when our proposed loss is used.

Subfields: Biomedical Engineering, Computer Vision and Pattern Recognition, Radiology, Nuclear Medicine and Imaging

View Paper: https://www.semanticscholar.org/paper/77e0d132a6c5da530b037c900cf326baa577dce3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biomedical Engineering

In Computer Vision and Pattern Recognition

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

A survey of loss functions for semantic segmentation (2020, background): “Combo loss [15] is defined as a weighted sum of Dice loss and a modified cross entropy.”
Unified Focal loss: Generalising Dice and cross entropy-based losses to handle class imbalanced medical image segmentation (2021, methodology): “To evaluate the performance stability of the γ hyperparameter, we display the DSC performance for each value of γ ∈ [0.1, 0.9] for the three datasets in Fig.”
MaX-DeepLab: End-to-End Panoptic Segmentation with Mask Transformers (2020, methodology): “This is consistent with the literature [42, 83, 10] that uses a cross-entropy loss together with a dice loss [69] to learn better segmentation masks.”
Medical Image Segmentation Using Deep Learning: A Survey (2020, background): “Besides, there are still a lot of works [77-78] trying to deal with this problem by adding penalties to loss functions or changing the optimization strategy according to specific tasks.”