skip to main content


Title: DOMINO: Domain-Aware Model Calibration in Medical Image Segmentation
Model calibration measures the agreement between the predicted probability estimates and the true correctness likelihood. Proper model calibration is vital for high-risk applications. Unfortunately, modern deep neural networks are poorly calibrated, compromising trustworthiness and reliability. Medical image segmentation particularly suffers from this due to the natural uncertainty of tissue boundaries. This is exasperated by their loss functions, which favor overconfidence in the majority classes. We address these challenges with DOMINO, a domain-aware model calibration method that leverages the semantic confusability and hierarchical similarity between class labels. Our experiments demonstrate that our DOMINO-calibrated deep neural networks outperform non-calibrated models and state-of-the-art morphometric methods in head image segmentation. Our results show that our method can consistently achieve better calibration, higher accuracy, and faster inference times than these methods, especially on rarer classes. This performance is attributed to our domain-aware regularization to inform semantic model calibration. These findings show the importance of semantic ties between class labels in building confidence in deep learning models. The framework has the potential to improve the trustworthiness and reliability of generic medical image segmentation models. The code for this article is available at: https://github.com/lab-smile/DOMINO.  more » « less
Award ID(s):
1908299
NSF-PAR ID:
10358376
Author(s) / Creator(s):
; ; ; ; ; ; ;
Editor(s):
Wang, L.; Dou, Q.; Fletcher, P.T.; Speidel, S.; Li, S.
Date Published:
Journal Name:
Medical Image Computing and Computer Assisted Intervention – MICCAI 2022
Volume:
13435
Page Range / eLocation ID:
454-463
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. Deep neural networks (DNNs) have started to find their role in the modern healthcare system. DNNs are being developed for diagnosis, prognosis, treatment planning, and outcome prediction for various diseases. With the increasing number of applications of DNNs in modern healthcare, their trustworthiness and reliability are becoming increasingly important. An essential aspect of trustworthiness is detecting the performance degradation and failure of deployed DNNs in medical settings. The softmax output values produced by DNNs are not a calibrated measure of model confidence. Softmax probability numbers are generally higher than the actual model confidence. The model confidence-accuracy gap further increases for wrong predictions and noisy inputs. We employ recently proposed Bayesian deep neural networks (BDNNs) to learn uncertainty in the model parameters. These models simultaneously output the predictions and a measure of confidence in the predictions. By testing these models under various noisy conditions, we show that the (learned) predictive confidence is well calibrated. We use these reliable confidence values for monitoring performance degradation and failure detection in DNNs. We propose two different failure detection methods. In the first method, we define a fixed threshold value based on the behavior of the predictive confidence with changing signal-to-noise ratio (SNR) of the test dataset. The second method learns the threshold value with a neural network. The proposed failure detection mechanisms seamlessly abstain from making decisions when the confidence of the BDNN is below the defined threshold and hold the decision for manual review. Resultantly, the accuracy of the models improves on the unseen test samples. We tested our proposed approach on three medical imaging datasets: PathMNIST, DermaMNIST, and OrganAMNIST, under different levels and types of noise. An increase in the noise of the test images increases the number of abstained samples. BDNNs are inherently robust and show more than 10% accuracy improvement with the proposed failure detection methods. The increased number of abstained samples or an abrupt increase in the predictive variance indicates model performance degradation or possible failure. Our work has the potential to improve the trustworthiness of DNNs and enhance user confidence in the model predictions. 
    more » « less
  2. Recent efforts in deploying Deep Neural Networks for object detection in real world applications, such as autonomous driving, assume that all relevant object classes have been observed during training. Quantifying the performance of these models in settings when the test data is not represented in the training set has mostly focused on pixel-level uncertainty estimation techniques of models trained for semantic segmentation. This paper proposes to exploit additional predictions of semantic segmentation models and quantifying its confidences, followed by classification of object hypotheses as known vs. unknown, out of distribution objects. We use object proposals generated by Region Proposal Network (RPN) and adapt distance aware uncertainty estimation of semantic segmentation using Radial Basis Functions Networks (RBFN) for class agnostic object mask prediction. The augmented object proposals are then used to train a classifier for known vs. unknown objects categories. Experimental results demonstrate that the proposed method achieves parallel performance to state of the art methods for unknown object detection and can also be used effectively for reducing object detectors' false positive rate. Our method is well suited for applications where prediction of non-object background categories obtained by semantic segmentation is reliable. 
    more » « less
  3. Recent years have witnessed the great success of deep learning models in semantic segmentation. Nevertheless, these models may not generalize well to unseen image domains due to the phenomenon of domain shift. Since pixel-level annotations are laborious to collect, developing algorithms which can adapt labeled data from source domain to target domain is of great significance. To this end, we propose self-ensembling attention networks to reduce the domain gap between different datasets. To the best of our knowledge, the proposed method is the first attempt to introduce selfensembling model to domain adaptation for semantic segmentation, which provides a different view on how to learn domain-invariant features. Besides, since different regions in the image usually correspond to different levels of domain gap, we introduce the attention mechanism into the proposed framework to generate attention-aware features, which are further utilized to guide the calculation of consistency loss in the target domain. Experiments on two benchmark datasets demonstrate that the proposed framework can yield competitive performance compared with the state of the art methods. 
    more » « less
  4. Given raster imagery features and imperfect vector training labels with registration uncertainty, this paper studies a deep learning framework that can quantify and reduce the registration uncertainty of training labels as well as train neural network parameters simultaneously. The problem is important in broad applications such as streamline classification on Earth imagery or tissue segmentation on medical imagery, whereby annotating precise vector labels is expensive and time-consuming. However, the problem is challenging due to the gap between the vector representation of class labels and the raster representation of image features and the need for training neural networks with uncertain label locations. Existing research on uncertain training labels often focuses on uncertainty in label class semantics or characterizes label registration uncertainty at the pixel level (not contiguous vectors). To fill the gap, this paper proposes a novel learning framework that explicitly quantifies vector labels' registration uncertainty. We propose a registration-uncertainty-aware loss function and design an iterative uncertainty reduction algorithm by re-estimating the posterior of true vector label locations distribution based on a Gaussian process. Evaluations on real-world datasets in National Hydrography Dataset refinement show that the proposed approach significantly outperforms several baselines in the registration uncertainty estimations performance and classification performance. 
    more » « less
  5. Given earth imagery with spectral features on a terrain surface, this paper studies surface segmentation based on both explanatory features and surface topology. The problem is important in many spatial and spatiotemporal applications such as flood extent mapping in hydrology. The problem is uniquely challenging for several reasons: first, the size of earth imagery on a terrain surface is often much larger than the input of popular deep convolutional neural networks; second, there exists topological structure dependency between pixel classes on the surface, and such dependency can follow an unknown and non-linear distribution; third, there are often limited training labels. Existing methods for earth imagery segmentation often divide the imagery into patches and consider the elevation as an additional feature channel. These methods do not fully incorporate the spatial topological structural constraint within and across surface patches and thus often show poor results, especially when training labels are limited. Existing methods on semi-supervised and unsupervised learning for earth imagery often focus on learning representation without explicitly incorporating surface topology. In contrast, we propose a novel framework that explicitly models the topological skeleton of a terrain surface with a contour tree from computational topology, which is guided by the physical constraint (e.g., water flow direction on terrains). Our framework consists of two neural networks: a convolutional neural network (CNN) to learn spatial contextual features on a 2D image grid, and a graph neural network (GNN) to learn the statistical distribution of physics-guided spatial topological dependency on the contour tree. The two models are co-trained via variational EM. Evaluations on the real-world flood mapping datasets show that the proposed models outperform baseline methods in classification accuracy, especially when training labels are limited. 
    more » « less