Search Results for author:
Found 55 papers, 29 papers with code
Privacy Distillation: Reducing Re-identification Risk of Multimodal Diffusion Models
Knowledge distillation in neural networks refers to compressing a large model or dataset into a smaller version of itself.
Parameter-Efficient Fine-Tuning for Medical Image Analysis: The Missed Opportunity
We present a comprehensive evaluation of Parameter-Efficient Fine-Tuning (PEFT) techniques for diverse medical image analysis tasks.
Understanding metric-related pitfalls in image analysis validation
Validation metrics are key for the reliable tracking of scientific progress and for bridging the current chasm between artificial intelligence (AI) research and its translation into practice.
The role of noise in denoising models for anomaly detection in medical images
Denoising methods, for instance classical denoising autoencoders (DAEs) and more recently emerging diffusion models, are a promising approach, however naive application of pixelwise noise leads to poor anomaly detection performance.
Clinically Plausible Pathology-Anatomy Disentanglement in Patient Brain MRI with Structured Variational Priors
We propose a hierarchically structured variational inference model for accurately disentangling observable evidence of disease (e. g. brain lesions or atrophy) from subject-specific anatomy in brain MRIs.
Rethinking Generalization: The Impact of Annotation Style on Medical Image Segmentation
This is particularly important in the context of medical image segmentation of pathological structures (e. g. lesions), where the annotation process is much more subjective, and affected by a number underlying factors, including the annotation protocol, rater education/experience, and clinical aims, among others.
Diffusion Models for Causal Discovery via Topological Ordering
Topological ordering approaches for causal discovery exploit this by performing graph discovery in two steps, first sequentially identifying nodes in reverse order of depth (topological ordering), and secondly pruning the potential relations.
HSIC-InfoGAN: Learning Unsupervised Disentangled Representations by Maximising Approximated Mutual Information
Maximisation of mutual information is achieved by introducing an auxiliary network and training with a latent regression loss.
What is Healthy? Generative Counterfactual Diffusion for Lesion Localization
This requires training with healthy and unhealthy data in DPMs.
vMFNet: Compositionality Meets Domain-generalised Segmentation
Moreover, with a reconstruction module, unlabeled data can also be used to learn the vMF kernels and likelihoods by recombining them to reconstruct the input image.
Why patient data cannot be easily forgotten?
With the advent of AI learned on data, one can imagine that such rights can extent to requests for forgetting knowledge of patient's data within AI models.
Metrics reloaded: Pitfalls and recommendations for image analysis validation
The framework was developed in a multi-stage Delphi process and is based on the novel concept of a problem fingerprint - a structured representation of the given problem that captures all aspects that are relevant for metric selection, from the domain interest to the properties of the target structure(s), data set and algorithm output.
Causal Machine Learning for Healthcare and Precision Medicine
Causal machine learning (CML) has experienced increasing popularity in healthcare.
Unintended memorisation of unique features in neural networks
Neural networks pose a privacy risk due to their propensity to memorise and leak training data.
Adversarial Counterfactual Augmentation: Application in Alzheimer's Disease Classification
To demonstrate the effectiveness of the proposed approach, we validate the method with the classification of Alzheimer's Disease (AD) as a downstream task.
Diffusion Causal Models for Counterfactual Estimation
We consider the task of counterfactual estimation from observational imaging data given a known causal structure.
Measuring Unintended Memorisation of Unique Private Features in Neural Networks
Neural networks pose a privacy risk to training data due to their propensity to memorise and leak information.
Indication as Prior Knowledge for Multimodal Disease Classification in Chest Radiographs with Transformers
We use the indication field to drive better image classification, by taking a transformer network which is unimodally pre-trained on text (BERT) and fine-tuning it for multimodal classification of a dual image-text input.
MyoPS: A Benchmark of Myocardial Pathology Segmentation Combining Three-Sequence Cardiac Magnetic Resonance Images
Assessment of myocardial viability is essential in diagnosis and treatment management of patients suffering from myocardial infarction, and classification of pathology on myocardium is the key to this assessment.
FROB: Few-shot ROBust Model for Classification and Out-of-Distribution Detection
By including our boundary, FROB reduces the threshold linked to the model's few-shot robustness; it maintains the OoD performance approximately independent of the number of few-shots.
OMASGAN: Out-of-Distribution Minimum Anomaly Score GAN for Sample Generation on the Boundary
OMASGAN addresses the rarity of anomalies by generating strong and adversarial OoD samples on the distribution boundary using only normal class data, effectively addressing mode collapse.
FROB: Few-shot ROBust Model for Classification with Out-of-Distribution Detection
We propose a self-supervised learning few-shot confidence boundary methodology based on generative and discriminative models, including classification.
Self-supervised Multi-scale Consistency for Weakly Supervised Segmentation Learning
Collecting large-scale medical datasets with fine-grained annotations is time-consuming and requires experts.
Stop Throwing Away Discriminators! Re-using Adversaries for Test-Time Training
At inference, the discriminator is discarded, and only the segmentor is used to predict label maps on test images.
Learning Disentangled Representations in the Imaging Domain
Disentangled representation learning has been proposed as an approach to learning general representations even in the absence of, or with limited, supervision.
Re-using Adversarial Mask Discriminators for Test-time Training under Distribution Shifts
After training is complete, the discriminator is usually discarded, and only the generator is used for inference.
Tail of Distribution GAN (TailGAN): Generative-Adversarial-Network-Based Boundary Formation
In this paper, we create a GAN-based tail formation model for anomaly detection, the Tail of distribution GAN (TailGAN), to generate samples on the tail of the data distribution and detect anomalies near the support boundary.
Boundary of Distribution Support Generator (BDSG): Sample Generation on the Boundary
We propose an invertible-residual-network-based model, the Boundary of Distribution Support Generator (BDSG).
Survey: Leakage and Privacy at Inference Time
Leakage of data from publicly available Machine Learning (ML) models is an area of growing significance as commercial and government applications of ML can draw on multiple sources of data, potentially including users' and clients' sensitive data.
Controllable cardiac synthesis via disentangled anatomy arithmetic
Motivated by the ability to disentangle images into spatial anatomy (tensor) factors and accompanying imaging (vector) representations, we propose a framework termed "disentangled anatomy arithmetic", in which a generative model learns to combine anatomical factors of different input images such that when they are re-entangled with the desired imaging modality (e. g. MRI), plausible new cardiac images are created with the target characteristics.
Semi-supervised Meta-learning with Disentanglement for Domain-generalised Medical Image Segmentation
We explicitly model the representations related to domain shifts.
Common Limitations of Image Processing Metrics: A Picture Story
While the importance of automatic image analysis is continuously increasing, recent meta-research revealed major flaws with respect to algorithm validation.
GaNDLF: A Generally Nuanced Deep Learning Framework for Scalable End-to-End Clinical Workflows in Medical Imaging
Deep Learning (DL) has the potential to optimize machine learning in both the scientific and clinical communities.
Max-Fusion U-Net for Multi-Modal Pathology Segmentation with Attention and Dynamic Resampling
Together with the corresponding encoding features, these representations are propagated to decoding layers with U-Net skip-connections.
Semi-supervised Pathology Segmentation with Disentangled Representations
The model is trained in a semi-supervised fashion with new reconstruction losses directly aiming to improve pathology segmentation with limited annotations.
Measuring the Biases and Effectiveness of Content-Style Disentanglement
In this paper, we conduct an empirical study to investigate the role of different biases in content-style disentanglement settings and unveil the relationship between the degree of disentanglement and task performance.
Disentangled Representations for Domain-generalized Cardiac Segmentation
Robust cardiac image segmentation is still an open challenge due to the inability of the existing methods to achieve satisfactory performance on unseen data of different domains.
INSIDE: Steering Spatial Attention with Non-Imaging Information in CNNs
We evaluate the method on two datasets: a new CLEVR-Seg dataset where we segment objects based on location, and the ACDC dataset conditioned on cardiac phase and slice location within the volume.
Learning to Segment from Scribbles using Multi-scale Adversarial Attention Gates
We evaluated our model on several medical (ACDC, LVSC, CHAOS) and non-medical (PPSS) datasets, and we report performance levels matching those achieved by models trained with fully annotated segmentation masks.
Have you forgotten? A method to assess if machine learning models have forgotten data
In the era of deep learning, aggregation of data from several sources is a common approach to ensuring data diversity.
Pseudo-healthy synthesis with pathology disentanglement and adversarial learning
In this paper, we present a model that is encouraged to disentangle the information of pathology from what seems to be healthy.
Teacher-Student chain for efficient semi-supervised histology image classification
Deep learning shows great potential for the domain of digital pathology.
Blind Inpainting of Large-scale Masks of Thin Structures with Adversarial and Reinforcement Learning
We evaluate our method in several datasets in medical imaging, plant science, and remote sensing.
Learning to synthesise the ageing brain without longitudinal data
Our method synthesises images conditioned on two factors: age (a continuous variable), and status of Alzheimer's Disease (AD, an ordinal variable).
Disentangle, align and fuse for multimodal and semi-supervised image segmentation
Core to our method is learning a disentangled decomposition into anatomical and imaging factors.
Temporal Consistency Objectives Regularize the Learning of Disentangled Representations
There has been an increasing focus in learning interpretable feature representations, particularly in applications such as medical image analysis that require explainability, whilst relying less on annotated data (since annotations can be tedious and costly).
FIRE: Unsupervised bi-directional inter-modality registration using deep networks
Inter-modality image registration is an critical preprocessing step for many applications within the routine clinical pathway.
Disentangled Representation Learning in Cardiac Image Analysis
We can venture further and consider that a medical image naturally factors into some spatial factors depicting anatomy and factors that denote the imaging characteristics.
The Generalized Complex Kernel Least-Mean-Square Algorithm
Also, the flexibility of the proposed generalized approach is tested in a second experiment with non-independent real and imaginary parts.
Adversarial Pseudo Healthy Synthesis Needs Pathology Factorization
Pseudo healthy synthesis, i. e. the creation of a subject-specific `healthy' image from a pathological one, could be helpful in tasks such as anomaly detection, understanding changes induced by pathology and disease or even as data augmentation.
Factorised spatial representation learning: application in semi-supervised myocardial segmentation
Specifically, we achieve comparable performance to fully supervised networks using a fraction of labelled images in experiments on ACDC and a dataset from Edinburgh Imaging Facility QMRI.
Leveraging multiple datasets for deep leaf counting
While state-of-the-art results on leaf counting with deep learning methods have recently been reported, they obtain the count as a result of leaf segmentation and thus require per-leaf (instance) segmentation to train the models (a rather strong annotation).
ARIGAN: Synthetic Arabidopsis Plants using Generative Adversarial Network
We show that our model is able to generate realistic 128x128 colour images of plants.
Theta-RBM: Unfactored Gated Restricted Boltzmann Machine for Rotation-Invariant Representations
In this paper, we propose the Theta-Restricted Boltzmann Machine ({\theta}-RBM in short), which builds upon the original RBM formulation and injects the notion of rotation-invariance during the learning procedure.
Rotation-Invariant Restricted Boltzmann Machine Using Shared Gradient Filters
Finding suitable features has been an essential problem in computer vision.
