Search Results for author:
Found 75 papers, 37 papers with code
Counterfactual Influence as a Distributional Quantity
Counterfactual self-influence is a popular metric to study memorization, quantifying how the model's prediction for a sample changes depending on the sample's inclusion in the training dataset.
The Hitchhiker's Guide to Efficient, End-to-End, and Tight DP Auditing
This paper systematizes research on auditing Differential Privacy (DP) techniques, aiming to identify key insights into the current state of the art and open challenges.
Laplace Sample Information: Data Informativeness Through a Bayesian Lens
Accurately estimating the informativeness of individual samples in a dataset is an important objective in deep learning, as it can guide sample selection, which can improve model efficiency and accuracy by removing redundant or potentially harmful samples.
$(\varepsilon, δ)$ Considered Harmful: Best Practices for Reporting Differential Privacy Guarantees
Current practices for reporting the level of differential privacy (DP) guarantees for machine learning (ML) algorithms provide an incomplete and potentially misleading picture of the guarantees and make it difficult to compare privacy levels across different settings.
Improved Localized Machine Unlearning Through the Lens of Memorization
We also propose a new unlearning algorithm, Deletion by Example Localization (DEL), that resets the parameters deemed-to-be most critical according to our localization strategy, and then finetunes them.
Differentially Private Active Learning: Balancing Effective Data Selection and Privacy
This work addresses this gap by introducing differentially private active learning (DP-AL) for standard learning settings.
On Differentially Private 3D Medical Image Synthesis with Controllable Latent Diffusion Models
Generally, the small size of public medical imaging datasets coupled with stringent privacy concerns, hampers the advancement of data-hungry deep learning models in medical imaging.
Enhancing the Utility of Privacy-Preserving Cancer Classification using Synthetic Data
This work addresses these challenges exploring and quantifying the utility of privacy-preserving deep learning techniques, concretely, (i) differentially private stochastic gradient descent (DP-SGD) and (ii) fully synthetic training data generated by our proposed malignancy-conditioned generative adversarial network.
Machine Unlearning for Medical Imaging
Machine unlearning is the process of removing the impact of a particular set of training samples from a pretrained model.
Attack-Aware Noise Calibration for Differential Privacy
For a given notion of attack risk, our approach significantly decreases noise scale, leading to increased utility at the same level of privacy.
Beyond the Calibration Point: Mechanism Comparison in Differential Privacy
In differentially private (DP) machine learning, the privacy guarantees of DP mechanisms are often reported and compared on the basis of a single $(\varepsilon, \delta)$-pair.
ChEX: Interactive Localization and Region Description in Chest X-rays
Report generation models offer fine-grained textual interpretations of medical images like chest X-rays, yet they often lack interactivity (i. e. the ability to steer the generation process through user queries) and localized interpretability (i. e. visually grounding their predictions), which we deem essential for future adoption in clinical practice.
Cross-domain and Cross-dimension Learning for Image-to-Graph Transformers
We propose (1) a regularized edge sampling loss to effectively learn object relations in multiple domains with different numbers of edges, (2) a domain adaptation framework for image-to-graph transformers aligning image- and graph-level features from different domains, and (3) a projection function that allows using 2D data for training 3D transformers.
Bounding Reconstruction Attack Success of Adversaries Without Data Priors
When training ML models with differential privacy (DP), formal upper bounds on the success of such reconstruction attacks can be provided.
Weakly Supervised Object Detection in Chest X-Rays with Differentiable ROI Proposal Networks and Soft ROI Pooling
Weakly supervised object detection (WSup-OD) increases the usefulness and interpretability of image classification algorithms without requiring additional supervision.
How Low Can You Go? Surfacing Prototypical In-Distribution Samples for Unsupervised Anomaly Detection
Generally, the assumption prevails that large training datasets allow the training of higher-performing UAD models.
Reconciling AI Performance and Data Reconstruction Resilience for Medical Imaging
Although a lower budget decreases the risk of information leakage, it typically also reduces the performance of such models.
SoK: Memorisation in machine learning
In this work we unify a broad range of previous definitions and perspectives on memorisation in ML, discuss their interplay with model generalisation and their implications of these phenomena on data privacy.
Propagation and Attribution of Uncertainty in Medical Imaging Pipelines
In this paper, we propose a method to propagate uncertainty through cascades of deep learning models in medical imaging pipelines.
(Predictable) Performance Bias in Unsupervised Anomaly Detection
The empirical fairness laws discovered in our study make disparate performance in UAD models easier to estimate and aid in determining the most desirable dataset composition.
MAD: Modality Agnostic Distance Measure for Image Registration
The success of multi-modal image registration, whether it is conventional or learning based, is predicated upon the choice of an appropriate distance (or similarity) measure.
Anatomy-Driven Pathology Detection on Chest X-rays
Pathology detection and delineation enables the automatic interpretation of medical scans such as chest X-rays while providing a high level of explainability to support radiologists in making informed decisions.
Bias-Aware Minimisation: Understanding and Mitigating Estimator Bias in Private SGD
Differentially private SGD (DP-SGD) holds the promise of enabling the safe and responsible application of machine learning to sensitive datasets.
FUTURE-AI: International consensus guideline for trustworthy and deployable artificial intelligence in healthcare
This work describes the FUTURE-AI guideline as the first international consensus framework for guiding the development and deployment of trustworthy AI tools in healthcare.
Extended Graph Assessment Metrics for Graph Neural Networks
In this work, we introduce extended graph assessment metrics (GAMs) for regression tasks and continuous adjacency matrices.
Body Fat Estimation from Surface Meshes using Graph Neural Networks
Body fat volume and distribution can be a strong indication for a person's overall health and the risk for developing diseases like type 2 diabetes and cardiovascular diseases.
Privacy-Utility Trade-offs in Neural Networks for Medical Population Graphs: Insights from Differential Privacy and Graph Structure
Our findings highlight the potential and the challenges of this specific DP application area.
Interpretable 2D Vision Models for 3D Medical Images
Training Artificial Intelligence (AI) models on 3D images presents unique challenges compared to the 2D case: Firstly, the demand for computational resources is significantly higher, and secondly, the availability of large datasets for pre-training is often limited, impeding training success.
Bounding data reconstruction attacks with the hypothesis testing interpretation of differential privacy
We explore Reconstruction Robustness (ReRo), which was recently proposed as an upper bound on the success of data reconstruction attacks against machine learning models.
Preserving privacy in domain transfer of medical AI models comes at no performance costs: The integral role of differential privacy
We specifically investigate the performance of models trained with DP as compared to models trained without DP on data from institutions that the model had not seen during its training (i. e., external validation) - the situation that is reflective of the clinical use of AI models.
Incentivising the federation: gradient-based metrics for data selection and valuation in private decentralised training
Obtaining high-quality data for collaborative training of machine learning models can be a challenging task due to A) regulatory concerns and B) a lack of data owner incentives to participate.
Interactive and Explainable Region-guided Radiology Report Generation
While previous methods generate reports without the possibility of human intervention and with limited explainability, our method opens up novel clinical use cases through additional interactive capabilities and introduces a high degree of transparency and explainability.
Ranked #1 on
Medical Report Generation
on MIMIC-CXR
Robust Detection Outcome: A Metric for Pathology Detection in Medical Images
To tackle this problem, we propose Robust Detection Outcome (RoDeO); a novel metric for evaluating algorithms for pathology detection in medical images, especially in chest X-rays.
Unsupervised Pathology Detection: A Deep Dive Into the State of the Art
Our experiments demonstrate that newly developed feature-modeling methods from the industrial and medical literature achieve increased performance compared to previous work and set the new SOTA in a variety of modalities and datasets.
Private, fair and accurate: Training large-scale, privacy-preserving AI models in medical imaging
In this work, we evaluated the effect of privacy-preserving training of AI models regarding accuracy and fairness compared to non-private training.
Equivariant Differentially Private Deep Learning: Why DP-SGD Needs Sparser Models
We achieve such sparsity by design by introducing equivariant convolutional networks for model training with Differential Privacy.
How Do Input Attributes Impact the Privacy Loss in Differential Privacy?
Differential privacy (DP) is typically formulated as a worst-case privacy guarantee over all individuals in a database.
The Role of Local Alignment and Uniformity in Image-Text Contrastive Learning on Medical Images
Image-text contrastive learning has proven effective for pretraining medical image models.
Exploiting segmentation labels and representation learning to forecast therapy response of PDAC patients
The prediction of pancreatic ductal adenocarcinoma therapy response is a clinically challenging and important task in this high-mortality tumour entity.
Generalised Likelihood Ratio Testing Adversaries through the Differential Privacy Lens
Differential Privacy (DP) provides tight upper bounds on the capabilities of optimal adversaries, but such adversaries are rarely encountered in practice.
Label Noise-Robust Learning using a Confidence-Based Sieving Strategy
Identifying the samples with noisy labels and preventing the model from learning them is a promising approach to address this challenge.
Kernel Normalized Convolutional Networks for Privacy-Preserving Machine Learning
They, on the other hand, considerably enhance the performance of shallow models in DP-FL and deeper models in FL and DP.
Bridging the Gap: Differentially Private Equivariant Deep Learning for Medical Image Analysis
Machine learning with formal privacy-preserving techniques like Differential Privacy (DP) allows one to derive valuable insights from sensitive medical imaging data while promising to protect patient privacy, but it usually comes at a sharp privacy-utility trade-off.
Unsupervised Anomaly Localization with Structural Feature-Autoencoders
Most commonly, the anomaly detection model generates a "normal" version of an input image, and the pixel-wise $l^p$-difference of the two is used to localize anomalies.
Kernel Normalized Convolutional Networks
Existing convolutional neural network architectures frequently rely upon batch normalization (BatchNorm) to effectively train the model.
SmoothNets: Optimizing CNN architecture design for differentially private deep learning
The arguably most widely employed algorithm to train deep neural networks with Differential Privacy is DPSGD, which requires clipping and noising of per-sample gradients.
Ranked #2 on
Image Classification
on Imagenette
Deep Learning
Image Classification with Differential Privacy
Can collaborative learning be private, robust and scalable?
In federated learning for medical image analysis, the safety of the learning protocol is paramount.
Relationformer: A Unified Framework for Image-to-Graph Generation
We leverage direct set-based object prediction and incorporate the interaction among the objects to learn an object-relation representation jointly.
SoK: Differential Privacy on Graph-Structured Data
In this work, we study the applications of differential privacy (DP) in the context of graph-structured data.
Beyond Gradients: Exploiting Adversarial Priors in Model Inversion Attacks
Collaborative machine learning settings like federated learning can be susceptible to adversarial interference and attacks.
Differentially private training of residual networks with scale normalisation
The training of neural networks with Differentially Private Stochastic Gradient Descent offers formal Differential Privacy guarantees but introduces accuracy trade-offs.
Multi-modal unsupervised brain image registration using edge maps
Diffeomorphic deformable multi-modal image registration is a challenging task which aims to bring images acquired by different modalities to the same coordinate space and at the same time to preserve the topology and the invertibility of the transformation.
On the Pitfalls of Using the Residual Error as Anomaly Score
Many current state-of-the-art methods for anomaly localization in medical images rely on calculating a residual image between a potentially anomalous input image and its "healthy" reconstruction.
Differentially Private Graph Classification with GNNs
In this work, we introduce differential privacy for graph-level classification, one of the key applications of machine learning on graphs.
AutoSeg -- Steering the Inductive Biases for Automatic Pathology Segmentation
In medical imaging, un-, semi-, or self-supervised pathology detection is often approached with anomaly- or out-of-distribution detection methods, whose inductive biases are not intentionally directed towards detecting pathologies, and are therefore sub-optimal for this task.
Distributed Machine Learning and the Semblance of Trust
The utilisation of large and diverse datasets for machine learning (ML) at scale is required to promote scientific insight into many meaningful problems.
Joint Learning of Localized Representations from Medical Images and Reports
Contrastive learning has proven effective for pre-training image models on unlabeled data with promising results for tasks such as medical image classification.
Complex-valued Federated Learning with Differential Privacy and MRI Applications
Experimentally, we showcase a proof-of-concept by training federated complex-valued neural networks with DP on a real-world task (MRI pulse sequence classification in $k$-space), yielding excellent utility and privacy.
An automatic differentiation system for the age of differential privacy
We introduce Tritium, an automatic differentiation-based sensitivity analysis framework for differentially private (DP) machine learning (ML).
Partial sensitivity analysis in differential privacy
However, while techniques such as individual R\'enyi DP (RDP) allow for granular, per-person privacy accounting, few works have investigated the impact of each input feature on the individual's privacy loss.
A unified interpretation of the Gaussian mechanism for differential privacy through the sensitivity index
$\psi$ uniquely characterises the GM and its properties by encapsulating its two fundamental quantities: the sensitivity of the query and the magnitude of the noise perturbation.
Challenging Current Semi-Supervised Anomaly Segmentation Methods for Brain MRI
In this work, we tackle the problem of Semi-Supervised Anomaly Segmentation (SAS) in Magnetic Resonance Images (MRI) of the brain, which is the task of automatically identifying pathologies in brain images.
Whole Brain Vessel Graphs: A Dataset and Benchmark for Graph Learning and Neuroscience (VesselGraph)
Moreover, we benchmark numerous state-of-the-art graph learning algorithms on the biologically relevant tasks of vessel prediction and vessel classification using the introduced vessel graph dataset.
NeuralDP Differentially private neural networks by design
The application of differential privacy to the training of deep neural networks holds the promise of allowing large-scale (decentralized) use of sensitive data while providing rigorous privacy guarantees to the individual.
Sensitivity analysis in differentially private machine learning using hybrid automatic differentiation
Reconciling large-scale ML with the closed-form reasoning required for the principled analysis of individual privacy loss requires the introduction of new tools for automatic sensitivity analysis and for tracking an individual's data and their features through the flow of computation.
Differentially private training of neural networks with Langevin dynamics for calibrated predictive uncertainty
We show that differentially private stochastic gradient descent (DP-SGD) can yield poorly calibrated, overconfident deep learning models.
Differentially private federated deep learning for multi-site medical image segmentation
The application of PTs to FL in medical imaging and the trade-offs between privacy guarantees and model utility, the ramifications on training performance and the susceptibility of the final models to attacks have not yet been conclusively investigated.
RATCHET: Medical Transformer for Chest X-ray Diagnosis and Reporting
Chest radiographs are one of the most common diagnostic modalities in clinical routine.
HyFed: A Hybrid Federated Framework for Privacy-preserving Machine Learning
Federated learning (FL) enables multiple clients to jointly train a global model under the coordination of a central server.
U-Noise: Learnable Noise Masks for Interpretable Image Segmentation
Deep Neural Networks (DNNs) are widely used for decision making in a myriad of critical applications, ranging from medical to societal and even judicial.
Privacy-preserving medical image analysis
The utilisation of artificial intelligence in medicine and healthcare has led to successful clinical applications in several domains.
Efficient, high-performance pancreatic segmentation using multi-scale feature extraction
For artificial intelligence-based image analysis methods to reach clinical applicability, the development of high-performance algorithms is crucial.
The Liver Tumor Segmentation Benchmark (LiTS)
In this work, we report the set-up and results of the Liver Tumor Segmentation Benchmark (LiTS), which was organized in conjunction with the IEEE International Symposium on Biomedical Imaging (ISBI) 2017 and the International Conferences on Medical Image Computing and Computer-Assisted Intervention (MICCAI) 2017 and 2018.
SurvivalNet: Predicting patient survival from diffusion weighted magnetic resonance images using cascaded fully convolutional and 3D convolutional neural networks
A 3D neural network (SurvivalNet) then predicts the HCC lesions' malignancy from the HCC tumor segmentation.
Automatic Liver and Tumor Segmentation of CT and MRI Volumes using Cascaded Fully Convolutional Neural Networks
In the first step, we train a FCN to segment the liver as ROI input for a second FCN.
Automatic Liver And Tumor Segmentation
Lesion Segmentation
+2
