Graph neural networks (GNNs) excel in graph representation learning by integrating graph structure and node features. Existing GNNs, unfortunately, fail to account for the uncertainty of class probabilities that vary with the depth of the model, leading to unreliable and risky predictions in real-world scenarios. To bridge the gap, in this paper, we propose a novel Evidence Fusing Graph Neural Network (EFGNN for short) to achieve trustworthy prediction, enhance node classification accuracy, and make explicit the risk of wrong predictions. In particular, we integrate the evidence theory with multi-hop propagation-based GNN architecture to quantify the prediction uncertainty of each node with the consideration of multiple receptive fields. Moreover, a parameter-free cumulative belief fusion (CBF) mechanism is developed to leverage the changes in prediction uncertainty and fuse the evidence to improve the trustworthiness of the final prediction. To effectively optimize the EFGNN model, we carefully design a joint learning objective composed of evidence cross-entropy, dissonance coefficient, and false confident penalty. The experimental results on various datasets and theoretical analyses demonstrate the effectiveness of the proposed model in terms of accuracy and trustworthiness, as well as its robustness to potential attacks. The source code of EFGNN is available at https://github.com/Shiy-Li/EFGNN.

Deep learning (DL) can automatically construct intelligent agents, deep neural networks (alternatively, DL models), that can outperform humans in certain tasks. However, the operating principles of DL remain poorly understood, making its decisions incomprehensible. As a result, it poses a great risk to deploy DL in high-stakes domains in which mistakes or errors may lead to critical consequences. Here, we aim to develop an algorithm that can help DL models make more robust decisions by allowing them to abstain from answering when they are uncertain. Our algorithm, named `Two out of Three (ToT)', is inspired by the sensitivity of the human brain to conflicting information. ToT creates two alternative predictions in addition to the original model prediction and uses the alternative predictions to decide whether it should provide an answer or not.

Purpose: This study introduces a novel Deep Learning (DL) - based q uality a sses s ment (QA) approach specifically designed for evaluating auto - generated contours (auto - contour s) in auto - segmentation for radiotherapy, with a focus on Online Adaptive Radiotherapy (OART). The proposed method leverages Bayesian Ordinal Classification (BOC), combined with cali brated thresholds derived from uncertainty quantification, to deliver confident QA predictions . This approach address es key challenges in clinical auto - segmentation QA workflows such as the absence of ground truth contours, limited availability of manually labeled data, and inherent uncertainty in AI model predictions . Methods: We developed a BOC model to classify the quality of auto - contour s and quantify uncertainty. To enhance predictive reliability, we implemented a calibration step to determine optimal uncertainty thresholds that meet specific clinical accuracy requirements . The method was validated under three distinct data availability scenarios: absence of manual labels, limited manual labeling, and extensive manual labeling. We specifically tested our method for auto - segmented rectum contours in prostate cancer radiotherapy. Geometric surrogate labels were employed in the absence of manual labels, transfer learning was applied when manual labels were limited, and direct use of manual labels was perf ormed when extensive labeling was available. Results: The BOC model demonstrated robust performance across all data availability scenarios for confident predictions, with significant accuracy gains when pre - trained with surrogate labels and fine - tuned with limited manual ly label ed data . Specifically, fine - tuning the pretrained model with just 30 manually labeled cases and calibrating with 34 subjects achieved over an accuracy of over 90% against manual labels in the test dataset .

Dropout and other feature noising schemes control overfitting by artificially corrupting the training data. For generalized linear models, dropout performs a form of adaptive regularization.

In real-world applications, one often encounters ambiguously labeled data, where different annotators assign conflicting class labels. Partial-label learning allows training classifiers in this weakly supervised setting. While state-of-the-art methods already feature good predictive performance, they often suffer from miscalibrated uncertainty estimates. However, having well-calibrated uncertainty estimates is important, especially in safety-critical domains like medicine and autonomous driving. In this article, we propose a novel nearest-neighbor-based partial-label-learning algorithm that leverages Dempster-Shafer theory. Extensive experiments on artificial and real-world datasets show that the proposed method provides a well-calibrated uncertainty estimate and achieves competitive prediction performance. Additionally, we prove that our algorithm is risk-consistent.

Deep learning has demonstrated remarkable achievements in medical image segmentation. However, prevailing deep learning models struggle with poor generalization due to (i) intra-class variations, where the same class appears differently in different samples, and (ii) inter-class independence, resulting in difficulties capturing intricate relationships between distinct objects, leading to higher false negative cases. This paper presents a novel approach that synergies spatial and spectral representations to enhance domain-generalized medical image segmentation. We introduce the innovative Spectral Correlation Coefficient objective to improve the model's capacity to capture middle-order features and contextual long-range dependencies. This objective complements traditional spatial objectives by incorporating valuable spectral information. Extensive experiments reveal that optimizing this objective with existing architectures like UNet and TransUNet significantly enhances generalization, interpretability, and noise robustness, producing more confident predictions. For instance, in cardiac segmentation, we observe a 0.81 pp and 1.63 pp (pp = percentage point) improvement in DSC over UNet and TransUNet, respectively. Our interpretability study demonstrates that, in most tasks, objectives optimized with UNet outperform even TransUNet by introducing global contextual information alongside local details. These findings underscore the versatility and effectiveness of our proposed method across diverse imaging modalities and medical domains.

The classification of electrocardiogram (ECG) plays a crucial role in the development of an automatic cardiovascular diagnostic system. However, considerable variances in ECG signals between individuals is a significant challenge. Changes in data distribution limit cross-domain utilization of a model. In this study, we propose a solution to classify ECG in an unlabeled dataset by leveraging knowledge obtained from labeled source domain. We present a domain-adaptive deep network based on cross-domain feature discrepancy optimization. Our method comprises three stages: pre-training, cluster-centroid computing, and adaptation. In pre-training, we employ a Distributionally Robust Optimization (DRO) technique to deal with the vanishing worst-case training loss. To enhance the richness of the features, we concatenate three temporal features with the deep learning features. The cluster computing stage involves computing centroids of distinctly separable clusters for the source using true labels, and for the target using confident predictions. We propose a novel technique to select confident predictions in the target domain. In the adaptation stage, we minimize compacting loss within the same cluster, separating loss across different clusters, inter-domain cluster discrepancy loss, and running combined loss to produce a domain-robust model. Experiments conducted in both cross-domain and cross-channel paradigms show the efficacy of the proposed method. Our method achieves superior performance compared to other state-of-the-art approaches in detecting ventricular ectopic beats (V), supraventricular ectopic beats (S), and fusion beats (F). Our method achieves an average improvement of 11.78% in overall accuracy over the non-domain-adaptive baseline method on the three test datasets.

Many clinical studies require the follow-up of patients over time. This is challenging: apart from frequently observed drop-out, there are often also organizational and financial challenges, which can lead to reduced data collection and, in turn, can complicate subsequent analyses. In contrast, there is often plenty of baseline data available of patients with similar characteristics and background information, e.g., from patients that fall outside the study time window. In this article, we investigate whether we can benefit from the inclusion of such unlabeled data instances to predict accurate survival times. In other words, we introduce a third level of supervision in the context of survival analysis, apart from fully observed and censored instances, we also include unlabeled instances. We propose three approaches to deal with this novel setting and provide an empirical comparison over fifteen real-life clinical and gene expression survival datasets. Our results demonstrate that all approaches are able to increase the predictive performance over independent test data. We also show that integrating the partial supervision provided by censored data in a semi-supervised wrapper approach generally provides the best results, often achieving high improvements, compared to not using unlabeled data.

We present DistillFlow, a knowledge distillation approach to learning optical flow. DistillFlow trains multiple teacher models and a student model, where challenging transformations are applied to the input of the student model to generate hallucinated occlusions as well as less confident predictions. Then, a self-supervised learning framework is constructed: confident predictions from teacher models are served as annotations to guide the student model to learn optical flow for those less confident predictions. The self-supervised learning framework enables us to effectively learn optical flow from unlabeled data, not only for non-occluded pixels, but also for occluded pixels. DistillFlow achieves state-of-the-art unsupervised learning performance on both KITTI and Sintel datasets. Our self-supervised pre-trained model also provides an excellent initialization for supervised fine-tuning, suggesting an alternate training paradigm in contrast to current supervised learning methods that highly rely on pre-training on synthetic data. At the time of writing, our fine-tuned models ranked 1st among all monocular methods on the KITTI 2015 benchmark, and outperform all published methods on the Sintel Final benchmark. More importantly, we demonstrate the generalization capability of DistillFlow in three aspects: framework generalization, correspondence generalization and cross-dataset generalization.

In many settings, unlabeled data is plentiful (think images, text, etc), while sufficient labeled data for supervised learning might be harder to obtain. In these situations, it can be difficult to determine how well the model will generalize. Most methods for assessing model performance rely on labeled data alone, e.g. Without enough labeled data these can be unreliable. Is there anything more we can learn about the model's ability to generalize from unlabeled data? In this article, I demonstrate how unlabeled data can frequently be used to bound test loss.