Beyond scalar losses: calibrating segmentation models via gradient vector field surgery

• URL: http://arxiv.org/abs/2607.14338v1
• Date: Wed, 15 Jul 2026 20:03:10 GMT
• Title: Beyond scalar losses: calibrating segmentation models via gradient vector field surgery
• Abstract summary: Region-based loss functions have established themselves as the de facto standard for highly class- and region-imbalanced segmentation tasks.<n>Models trained using region-based loss functions are notoriously miscalibrated and typically yield over-confident predictions.
• Score: 40.96272223539153
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Region-based loss functions, such as the Dice loss, have established themselves as the de facto standard for highly class- and region-imbalanced segmentation tasks. However, models trained using region-based loss functions are notoriously miscalibrated and typically yield over-confident predictions. In medical imaging applications, such as defining tumor resection margins, this miscalibration is hindering clinical adoption. In this work, we outline a novel gradient perspective on this overconfidence and show how it affects region-based loss functions. We propose a "surgery" on the gradient vector field as a simple, yet effective intervention to mitigate calibration issues. This surgery adds a factor to the loss's partial derivative, scaling the gradient's magnitude linearly with the prediction error. In empirical evaluations across 2D and 3D medical segmentation tasks, we demonstrate the effectiveness of this intervention while maintaining high prediction accuracy when used in conjunction with any region-based loss function.

Related papers

• Medical Image Segmentation based on Deep Active Contour and Mean Curvature Loss Function [3.8546485293265977]
  We propose a loss function where the convolution kernel is used to approximate the mean curvature to save computational cost.<n>We have validated the performance of our method on the liver and spleen dataset.
  arXiv  Detail & Related papers  (2026-07-14T09:56:03Z)
• Orthogonal Causal Calibration [55.28164682911196]
  We develop general algorithms for reducing the task of causal calibration to that of calibrating a standard (non-causal) predictive model.<n>Our results are exceedingly general, showing that essentially any existing calibration algorithm can be used in causal settings.
  arXiv  Detail & Related papers  (2024-06-04T03:35:25Z)
• Mitigating False Predictions In Unreasonable Body Regions [0.921264855324451]
  We propose a novel loss function that penalizes predictions in implausible body regions. It is realized with a Body Part Regression model that generates axial slice positional scores. It effectively mitigates false positive tumor predictions up to 85% and significantly enhances overall segmentation performance.
  arXiv  Detail & Related papers  (2024-04-24T08:11:18Z)
• Cross-Dataset Adaptation for Instrument Classification in Cataract Surgery Videos [54.1843419649895]
  State-of-the-art models, which perform this task well on a particular dataset, perform poorly when tested on another dataset. We propose a novel end-to-end Unsupervised Domain Adaptation (UDA) method called the Barlow Adaptor. In addition, we introduce a novel loss called the Barlow Feature Alignment Loss (BFAL) which aligns features across different domains.
  arXiv  Detail & Related papers  (2023-07-31T18:14:18Z)
• Bridging Precision and Confidence: A Train-Time Loss for Calibrating Object Detection [58.789823426981044]
  We propose a novel auxiliary loss formulation that aims to align the class confidence of bounding boxes with the accurateness of predictions. Our results reveal that our train-time loss surpasses strong calibration baselines in reducing calibration error for both in and out-domain scenarios.
  arXiv  Detail & Related papers  (2023-03-25T08:56:21Z)
• A Generalized Surface Loss for Reducing the Hausdorff Distance in Medical Imaging Segmentation [1.2289361708127877]
  We propose a novel loss function to minimize Hausdorff-based metrics with more desirable numerical properties than current methods. Our loss function outperforms other losses when tested on the LiTS and BraTS datasets using the state-of-the-art nnUNet architecture.
  arXiv  Detail & Related papers  (2023-02-08T04:01:42Z)
• Calibrating Segmentation Networks with Margin-based Label Smoothing [19.669173092632]
  We provide a unifying constrained-optimization perspective of current state-of-the-art calibration losses. These losses could be viewed as approximations of a linear penalty imposing equality constraints on logit distances. We propose a simple and flexible generalization based on inequality constraints, which imposes a controllable margin on logit distances.
  arXiv  Detail & Related papers  (2022-09-09T20:21:03Z)
• Self-Supervised Training with Autoencoders for Visual Anomaly Detection [61.62861063776813]
  We focus on a specific use case in anomaly detection where the distribution of normal samples is supported by a lower-dimensional manifold. We adapt a self-supervised learning regime that exploits discriminative information during training but focuses on the submanifold of normal examples. We achieve a new state-of-the-art result on the MVTec AD dataset -- a challenging benchmark for visual anomaly detection in the manufacturing domain.
  arXiv  Detail & Related papers  (2022-06-23T14:16:30Z)
• A novel shape-based loss function for machine learning-based seminal organ segmentation in medical imaging [1.0312968200748116]
  Deep convolutional neural networks have exhibited promising performance in accurate and automatic seminal segmentation. A novel shape-based cost function is proposed which encourages/constrains the network to learn/capture the underlying shape features.
  arXiv  Detail & Related papers  (2022-03-07T12:26:30Z)
• Anatomical and Diagnostic Bayesian Segmentation in Prostate MRI $-$Should Different Clinical Objectives Mandate Different Loss Functions? [0.3149883354098941]
  Probable voxel-level classification of anatomy and malignancy in prostate MRI requires different loss functions for optimal performance. We investigate distribution, region and boundary-based loss functions for both tasks across 200 patient exams from the publicly-available ProstateX dataset.
  arXiv  Detail & Related papers  (2021-10-25T12:44:52Z)
• An Uncertainty-Driven GCN Refinement Strategy for Organ Segmentation [53.425900196763756]
  We propose a segmentation refinement method based on uncertainty analysis and graph convolutional networks. We employ the uncertainty levels of the convolutional network in a particular input volume to formulate a semi-supervised graph learning problem. We show that our method outperforms the state-of-the-art CRF refinement method by improving the dice score by 1% for the pancreas and 2% for spleen.
  arXiv  Detail & Related papers  (2020-12-06T18:55:07Z)
• Weakly-supervised Learning For Catheter Segmentation in 3D Frustum Ultrasound [74.22397862400177]
  We propose a novel Frustum ultrasound based catheter segmentation method. The proposed method achieved the state-of-the-art performance with an efficiency of 0.25 second per volume.
  arXiv  Detail & Related papers  (2020-10-19T13:56:22Z)

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.