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

arXiv: https://arxiv.org/abs/1805.02798 · HTML: https://mlsanity.com/paper/1805.02798/

## Key facts

- Cited by 76 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 May 2018.

## Built on by (most cited first, 30 of 76)

- [Unified Focal loss: Generalising Dice and cross entropy-based losses to handle class imbalanced medical image segmentation](https://mlsanity.com/paper/2102.04525/) (2021)
- [Loss Functions in the Era of Semantic Segmentation: A Survey and Outlook](https://mlsanity.com/paper/2312.05391/) (2023)
- [Learning General-Purpose Biomedical Volume Representations using Randomized Synthesis](https://mlsanity.com/paper/2411.02372/) (2024)
- [Deep Learning methods for automatic evaluation of delayed enhancement-MRI. The results of the EMIDEC challenge](https://mlsanity.com/paper/2108.04016/) (2021)
- [Centerline Boundary Dice Loss for Vascular Segmentation](https://mlsanity.com/paper/2407.01517/) (2024)
- [MonoTrack: Shuttle trajectory reconstruction from monocular badminton video](https://mlsanity.com/paper/2204.01899/) (2022)
- [Ground then Navigate: Language-guided Navigation in Dynamic Scenes](https://mlsanity.com/paper/2209.11972/) (2022)
- [Striking the Right Balance: Recall Loss for Semantic Segmentation](https://mlsanity.com/paper/2106.14917/) (2021)
- [Y-Net: A Spatiospectral Dual-Encoder Networkfor Medical Image Segmentation](https://mlsanity.com/paper/2204.07613/) (2022)
- [Calibrating the Dice loss to handle neural network overconfidence for biomedical image segmentation](https://mlsanity.com/paper/2111.00528/) (2021)
- [A Generalized Surface Loss for Reducing the Hausdorff Distance in Medical Imaging Segmentation](https://mlsanity.com/paper/2302.03868/) (2023)
- [M2SNet: Multi-scale in Multi-scale Subtraction Network for Medical Image Segmentation](https://mlsanity.com/paper/2303.10894/) (2023)
- [Towards automated brain aneurysm detection in TOF-MRA: open data, weak labels, and anatomical knowledge](https://mlsanity.com/paper/2103.06168/) (2021)
- [Do We Really Need Dice? The Hidden Region-Size Biases of Segmentation Losses](https://mlsanity.com/paper/2104.08717/) (2021)
- [A Comprehensive Survey on Imbalanced Data Learning](https://mlsanity.com/paper/2502.08960/) (2025)
- [Multi-site, Multi-domain Airway Tree Modeling (ATM'22): A Public Benchmark for Pulmonary Airway Segmentation](https://mlsanity.com/paper/2303.05745/) (2023)
- [MDViT: Multi-domain Vision Transformer for Small Medical Image Segmentation Datasets](https://mlsanity.com/paper/2307.02100/) (2023)
- [Widely Applicable Strong Baseline for Sports Ball Detection and Tracking](https://mlsanity.com/paper/2311.05237/) (2023)
- [Region-wise Loss for Biomedical Image Segmentation](https://mlsanity.com/paper/2108.01405/) (2021)
- [Joint Learning of Blind Super-Resolution and Crack Segmentation for Realistic Degraded Images](https://mlsanity.com/paper/2302.12491/) (2023)
- [LATUP-Net: A Lightweight 3D Attention U-Net with Parallel Convolutions for Brain Tumor Segmentation](https://mlsanity.com/paper/2404.05911/) (2024)
- [A Data-Adaptive Loss Function for Incomplete Data and Incremental Learning in Semantic Image Segmentation](https://mlsanity.com/paper/2104.11020/) (2021)
- [Beyond Accuracy Optimization: Computer Vision Losses for Large Language Model Fine-Tuning](https://mlsanity.com/paper/2409.13641/) (2024)
- [Hypernet-Ensemble Learning of Segmentation Probability for Medical Image Segmentation with Ambiguous Labels](https://mlsanity.com/paper/2112.06693/) (2021)
- [DOMINO: Domain-aware Model Calibration in Medical Image Segmentation](https://mlsanity.com/paper/2209.06077/) (2022)
- [Shape-Sensitive Loss for Catheter and Guidewire Segmentation](https://mlsanity.com/paper/2311.11205/) (2023)
- [Building a Synthetic Vascular Model: Evaluation in an Intracranial Aneurysms Detection Scenario](https://mlsanity.com/paper/2411.02477/) (2024)
- [GATE3D: Generalized Attention-based Task-synergized Estimation in 3D*](https://mlsanity.com/paper/2504.11014/) (2025)
- [SemSegLoss: A python package of loss functions for semantic segmentation](https://mlsanity.com/paper/2106.05844/) (2021)
- [Adaptively Weighted Data Augmentation Consistency Regularization for Robust Optimization under Concept Shift](https://mlsanity.com/paper/2210.01891/) (2022)

## Abstract (arXiv)

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.