Tingting Zhu

Shuo Zhang, Chenqi Li, Tingting Zhu

Long-tailed recognition poses a significant challenge for deep learning. The two-stage decoupling paradigm, which separates representation learning from classifier retraining, offers a promising solution. During the classifier retraining stage, adaptive norm rescaling is a popular technique. It adjusts the per-class weight norms via parameter regularization, which inevitably introduces hyperparameters. However, many studies report that long-tailed recognition is sensitive to these hyperparameters, as their setup significantly impacts performance. In this paper, we first provide a class-conditional distribution perspective to support norm rescaling methods. Furthermore, we propose a simple but effective approach called Self-Adaptive Monotonic Normalization (SAMN). SAMN avoids the need for parameter regularization. It directly enforces monotonicity on per-class weight norms using the Pool Adjacent Violators Algorithm, making the method hyperparameter-friendly. SAMN is a universal strategy that integrates seamlessly with other methods for enhanced performance. Experiments on benchmark datasets demonstrate that our method significantly boosts long-tailed recognition performance, often achieving state-of-the-art results.

Vinod Kumar Chauhan, Soheila Molaei, Marzia Hoque Tania et al. · oxford

Observational studies have recently received significant attention from the machine learning community due to the increasingly available non-experimental observational data and the limitations of the experimental studies, such as considerable cost, impracticality, small and less representative sample sizes, etc. In observational studies, de-confounding is a fundamental problem of individualised treatment effects (ITE) estimation. This paper proposes disentangled representations with adversarial training to selectively balance the confounders in the binary treatment setting for the ITE estimation. The adversarial training of treatment policy selectively encourages treatment-agnostic balanced representations for the confounders and helps to estimate the ITE in the observational studies via counterfactual inference. Empirical results on synthetic and real-world datasets, with varying degrees of confounding, prove that our proposed approach improves the state-of-the-art methods in achieving lower error in the ITE estimation.

Ghadeer Ghosheh, Jin Li, Tingting Zhu

Electronic Health Records (EHRs) are a valuable asset to facilitate clinical research and point of care applications; however, many challenges such as data privacy concerns impede its optimal utilization. Deep generative models, particularly, Generative Adversarial Networks (GANs) show great promise in generating synthetic EHR data by learning underlying data distributions while achieving excellent performance and addressing these challenges. This work aims to review the major developments in various applications of GANs for EHRs and provides an overview of the proposed methodologies. For this purpose, we combine perspectives from healthcare applications and machine learning techniques in terms of source datasets and the fidelity and privacy evaluation of the generated synthetic datasets. We also compile a list of the metrics and datasets used by the reviewed works, which can be utilized as benchmarks for future research in the field. We conclude by discussing challenges in GANs for EHRs development and proposing recommended practices. We hope that this work motivates novel research development directions in the intersection of healthcare and machine learning.

Munib Mesinovic, Peter Watkinson, Tingting Zhu

Recent advancements in AI applications to healthcare have shown incredible promise in surpassing human performance in diagnosis and disease prognosis. With the increasing complexity of AI models, however, concerns regarding their opacity, potential biases, and the need for interpretability. To ensure trust and reliability in AI systems, especially in clinical risk prediction models, explainability becomes crucial. Explainability is usually referred to as an AI system's ability to provide a robust interpretation of its decision-making logic or the decisions themselves to human stakeholders. In clinical risk prediction, other aspects of explainability like fairness, bias, trust, and transparency also represent important concepts beyond just interpretability. In this review, we address the relationship between these concepts as they are often used together or interchangeably. This review also discusses recent progress in developing explainable models for clinical risk prediction, highlighting the importance of quantitative and clinical evaluation and validation across multiple common modalities in clinical practice. It emphasizes the need for external validation and the combination of diverse interpretability methods to enhance trust and fairness. Adopting rigorous testing, such as using synthetic datasets with known generative factors, can further improve the reliability of explainability methods. Open access and code-sharing resources are essential for transparency and reproducibility, enabling the growth and trustworthiness of explainable research. While challenges exist, an end-to-end approach to explainability in clinical risk prediction, incorporating stakeholders from clinicians to developers, is essential for success.

Taha Ceritli, Ghadeer O. Ghosheh, Vinod Kumar Chauhan et al.

Electronic Health Records (EHRs) contain sensitive patient information, which presents privacy concerns when sharing such data. Synthetic data generation is a promising solution to mitigate these risks, often relying on deep generative models such as Generative Adversarial Networks (GANs). However, recent studies have shown that diffusion models offer several advantages over GANs, such as generation of more realistic synthetic data and stable training in generating data modalities, including image, text, and sound. In this work, we investigate the potential of diffusion models for generating realistic mixed-type tabular EHRs, comparing TabDDPM model with existing methods on four datasets in terms of data quality, utility, privacy, and augmentation. Our experiments demonstrate that TabDDPM outperforms the state-of-the-art models across all evaluation metrics, except for privacy, which confirms the trade-off between privacy and utility.

Munib Mesinovic, Tingting Zhu

Electronic health record (EHR) data present tremendous opportunities for advancing survival analysis through deep learning, yet reproducibility remains severely constrained by inconsistent preprocessing methodologies. We present SurvBench, a comprehensive, open-source preprocessing pipeline that transforms raw PhysioNet datasets into standardised, model-ready tensors for multi-modal survival analysis. SurvBench provides data loaders for three major critical care databases, MIMIC-IV, eICU, and MC-MED, supporting diverse modalities including time-series vitals, static demographics, ICD diagnosis codes, and radiology reports. The pipeline implements rigorous data quality controls, patient-level splitting to prevent data leakage, explicit missingness tracking, and standardised temporal aggregation. SurvBench handles both single-risk (e.g., in-hospital mortality) and competing-risks scenarios (e.g., multiple discharge outcomes). The outputs are compatible with pycox library packages and implementations of standard statistical and deep learning models. By providing reproducible, configuration-driven preprocessing with comprehensive documentation, SurvBench addresses the "preprocessing gap" that has hindered fair comparison of deep learning survival models, enabling researchers to focus on methodological innovation rather than data engineering.

Runlong Ye, Naaz Sibia, Angela Zavaleta Bernuy et al. · utoronto

Systematic Literature Reviews (SLRs) are fundamental to scientific progress, yet the process is hindered by a fragmented tool ecosystem that imposes a high cognitive load. This friction suppresses the iterative, exploratory nature of scholarly work. To investigate these challenges, we conducted an exploratory design study with 20 experienced researchers. This study identified key friction points: 1) the high cognitive load of managing iterative query refinement across multiple databases, 2) the overwhelming scale and pace of publication of modern literature, and 3) the tension between automation and scholarly agency. Informed by these findings, we developed ARC, a design probe that operationalizes solutions for multi-database integration, transparent iterative search, and verifiable AI-assisted screening. A comparative user study with 8 researchers suggests that an integrated environment facilitates a transition in scholarly work, moving researchers from managing administrative overhead to engaging in strategic exploration. By utilizing external representations to scaffold strategic exploration and transparent AI reasoning, our system supports verifiable judgment, aiming to augment expert contributions from initial creation through long-term maintenance of knowledge synthesis.

Tingting Zhu, Bo Peng, Jifan Liang et al.

Semantic communications (SC) have been expected to be a new paradigm shifting to catalyze the next generation communication, whose main concerns shift from accurate bit transmission to effective semantic information exchange in communications. However, the previous and widely-used metrics for images are not applicable to evaluate the image semantic similarity in SC. Classical metrics to measure the similarity between two images usually rely on the pixel level or the structural level, such as the PSNR and the MS-SSIM. Straightforwardly using some tailored metrics based on deep-learning methods in CV community, such as the LPIPS, is infeasible for SC. To tackle this, inspired by BERTScore in NLP community, we propose a novel metric for evaluating image semantic similarity, named Vision Transformer Score (ViTScore). We prove theoretically that ViTScore has 3 important properties, including symmetry, boundedness, and normalization, which make ViTScore convenient and intuitive for image measurement. To evaluate the performance of ViTScore, we compare ViTScore with 3 typical metrics (PSNR, MS-SSIM, and LPIPS) through 4 classes of experiments: (i) correlation with BERTScore through evaluation of image caption downstream CV task, (ii) evaluation in classical image communications, (iii) evaluation in image semantic communication systems, and (iv) evaluation in image semantic communication systems with semantic attack. Experimental results demonstrate that ViTScore is robust and efficient in evaluating the semantic similarity of images. Particularly, ViTScore outperforms the other 3 typical metrics in evaluating the image semantic changes by semantic attack, such as image inverse with Generative Adversarial Networks (GANs). This indicates that ViTScore is an effective performance metric when deployed in SC scenarios.

Munib Mesinovic, Peter Watkinson, Tingting Zhu

Machine learning applications for longitudinal electronic health records often forecast the risk of events at fixed time points, whereas survival analysis achieves dynamic risk prediction by estimating time-to-event distributions. Here, we propose a novel conditional variational autoencoder-based method, DySurv, which uses a combination of static and longitudinal measurements from electronic health records to estimate the individual risk of death dynamically. DySurv directly estimates the cumulative risk incidence function without making any parametric assumptions on the underlying stochastic process of the time-to-event. We evaluate DySurv on 6 time-to-event benchmark datasets in healthcare, as well as 2 real-world intensive care unit (ICU) electronic health records (EHR) datasets extracted from the eICU Collaborative Research (eICU) and the Medical Information Mart for Intensive Care database (MIMIC-IV). DySurv outperforms other existing statistical and deep learning approaches to time-to-event analysis across concordance and other metrics. It achieves time-dependent concordance of over 60% in the eICU case. It is also over 12% more accurate and 22% more sensitive than in-use ICU scores like Acute Physiology and Chronic Health Evaluation (APACHE) and Sequential Organ Failure Assessment (SOFA) scores. The predictive capacity of DySurv is consistent and the survival estimates remain disentangled across different datasets. Our interdisciplinary framework successfully incorporates deep learning, survival analysis, and intensive care to create a novel method for time-to-event prediction from longitudinal health records. We test our method on several held-out test sets from a variety of healthcare datasets and compare it to existing in-use clinical risk scoring benchmarks.

Zhikang Chen, Yue Wang, Sen Cui et al.

Electrocardiogram (ECG)-based models have achieved strong performance in diagnostic tasks, yet they remain limited in modeling how cardiac dynamics evolve under external interventions. In particular, existing approaches focus primarily on static prediction and lack mechanisms to capture ECG variations under different pharmacological conditions. In this work, we propose an ECG World Model for action-conditioned predictive simulation of cardiac electrophysiology. Moving beyond disjoint pipelines, our framework features a principled integration of physiological ordinary differential equation (ODE) priors into latent diffusion dynamics via energy regularization. This structural constraint enables the synthesis of physiologically plausible post-intervention ECG trajectories while effectively mitigating generative hallucinations. Building on this simulation process, we introduce an uncertainty-aware evaluation strategy that leverages the stochasticity of diffusion sampling to characterize both the expected clinical risk and its variability, allowing a more reliable comparative assessment of candidate interventions. We evaluate our method across diverse settings, including controlled drug-response scenarios and real-world clinical records. Beyond standard waveform metrics, experimental results demonstrate improved risk calibration and strong alignment with expert-informed treatment preferences. These results establish our approach as a robust foundation for safe and intervention-aware clinical decision support.

Jianqiao Chen, Nan Ma, Xiaodong Xu et al.

Digital mapping of semantic features is essential for achieving interoperability between semantic communication and practical digital infrastructure. However, current research efforts predominantly concentrate on analog semantic communication with simplified channel models. To bridge these gaps, we develop a robust vector quantized-enabled digital semantic communication (VQ-DSC-R) system built upon orthogonal frequency division multiplexing (OFDM) transmission. Our work encompasses the framework design of VQ-DSC-R, followed by a comprehensive optimization study. Firstly, we design a Swin Transformer-based backbone for hierarchical semantic feature extraction, integrated with VQ modules that map the features into a shared semantic quantized codebook (SQC) for efficient index transmission. Secondly, we propose a differentiable vector quantization with adaptive noise-variance (ANDVQ) scheme to mitigate quantization errors in SQC, which dynamically adjusts the quantization process using K-nearest neighbor statistics, while exponential moving average mechanism stabilizes SQC training. Thirdly, for robust index transmission over multipath fading channel and noise, we develop a conditional diffusion model (CDM) to refine channel state information, and design an attention-based module to dynamically adapt to channel noise. The entire VQ-DSC-R system is optimized via a three-stage training strategy. Extensive experiments demonstrate superiority of VQ-DSC-R over benchmark schemes, achieving high compression ratios and robust performance in practical scenarios.

Mingcheng Zhu, Zhiyao Luo, Yu Liu et al.

By processing electronic health records (EHRs) as natural language sequences, large language models (LLMs) have shown potential in clinical prediction tasks such as mortality prediction and phenotyping. However, longitudinal or highly frequent EHRs often yield excessively long token sequences that result in high computational costs and even reduced performance. Existing solutions either add modules for compression or remove less important tokens, which introduce additional inference latency or risk losing clinical information. To achieve lossless compression of token sequences without additional cost or loss of performance, we propose Medical Token-Pair Encoding (MedTPE), a layered method that extends standard tokenisation for EHR sequences. MedTPE merges frequently co-occurring medical token pairs into composite tokens, providing lossless compression while preserving the computational complexity through a dependency-aware replacement strategy. Only the embeddings of the newly introduced tokens of merely 0.5-1.0% of the LLM's parameters are fine-tuned via self-supervised learning. Experiments on real-world datasets for two clinical scenarios demonstrate that MedTPE reduces input token length by up to 31% and inference latency by 34-63%, while maintaining or even improving both predictive performance and output format compliance across multiple LLMs and four clinical prediction tasks. Furthermore, MedTPE demonstrates robustness across different input context lengths and generalisability to scientific and financial domains and different languages.

Munib Mesinovic, Max Buhlan, Tingting Zhu

Healthcare artificial intelligence systems routinely fail when deployed across institutions, with documented performance drops and perpetuation of discriminatory patterns embedded in historical data. This brittleness stems, in part, from learning statistical associations rather than causal mechanisms. Causal graph neural networks address this triple crisis of distribution shift, discrimination, and inscrutability by combining graph-based representations of biomedical data with causal inference principles to learn invariant mechanisms rather than spurious correlations. This Review examines methodological foundations spanning structural causal models, disentangled causal representation learning, and techniques for interventional prediction and counterfactual reasoning on graphs. We analyse applications demonstrating clinical value across psychiatric diagnosis through brain network analysis, cancer subtyping via multi-omics causal integration, continuous physiological monitoring with mechanistic interpretation, and drug recommendation correcting prescription bias. These advances establish foundations for patient-specific Causal Digital Twins, enabling in silico clinical experimentation, with integration of large language models for hypothesis generation and causal graph neural networks for mechanistic validation. Substantial barriers remain, including computational requirements precluding real-time deployment, validation challenges demanding multi-modal evidence triangulation beyond cross-validation, and risks of causal-washing where methods employ causal terminology without rigorous evidentiary support. We propose tiered frameworks distinguishing causally-inspired architectures from causally-validated discoveries and identify critical research priorities making causal rather than purely associational claims.

Zhiyao Luo, Tingting Zhu

Reinforcement learning (RL)-based dynamic treatment regimes (DTRs) hold promise for automating complex clinical decision-making, yet their practical deployment remains hindered by the intensive engineering required to inject clinical knowledge and ensure patient safety. Recent advancements in large language models (LLMs) suggest a complementary approach, where implicit prior knowledge and clinical heuristics are naturally embedded through linguistic prompts without requiring environment-specific training. In this study, we rigorously evaluate open-source LLMs as dynamic insulin dosing agents in an in silico Type 1 diabetes simulator, comparing their zero-shot inference performance against small neural network-based RL agents (SRAs) explicitly trained for the task. Our results indicate that carefully designed zero-shot prompts enable smaller LLMs (e.g., Qwen2.5-7B) to achieve comparable or superior clinical performance relative to extensively trained SRAs, particularly in stable patient cohorts. However, LLMs exhibit notable limitations, such as overly aggressive insulin dosing when prompted with chain-of-thought (CoT) reasoning, highlighting critical failure modes including arithmetic hallucination, temporal misinterpretation, and inconsistent clinical logic. Incorporating explicit reasoning about latent clinical states (e.g., meals) yielded minimal performance gains, underscoring the current model's limitations in capturing complex, hidden physiological dynamics solely through textual inference. Our findings advocate for cautious yet optimistic integration of LLMs into clinical workflows, emphasising the necessity of targeted prompt engineering, careful validation, and potentially hybrid approaches that combine linguistic reasoning with structured physiological modelling to achieve safe, robust, and clinically effective decision-support systems.

Mingcheng Zhu, Yu Liu, Zhiyao Luo et al.

The widespread adoption of electronic health records (EHRs) enables the acquisition of heterogeneous clinical data, spanning lab tests, vital signs, medications, and procedures, which offer transformative potential for artificial intelligence in healthcare. Although traditional modelling approaches have typically relied on multivariate time series, they often struggle to accommodate the inherent sparsity and irregularity of real-world clinical workflows. Consequently, research has shifted toward event stream representation, which treats patient records as continuous sequences, thereby preserving the precise temporal structure of the patient journey. However, the existing literature remains fragmented, characterised by inconsistent definitions, disparate modelling architectures, and varying training protocols. To address these gaps, this review establishes a unified definition of EHR event streams and introduces a novel taxonomy that categorises models based on their handling of event time, type, and value. We systematically review training strategies, ranging from supervised learning to self-supervised methods, and provide a comprehensive discussion of applications across clinical scenarios. Finally, we identify open critical challenges and future directions, with the aim of clarifying the current landscape and guiding the development of next-generation healthcare models.

Mingcheng Zhu, Yu Liu, Tingting Zhu

Uncertainty estimation is important for deploying LLMs in high-stakes applications such as healthcare and finance, where hallucinations can appear fluent and plausible while being factually incorrect, making it difficult for users to judge whether an output should be trusted. Existing methods require one or more full autoregressive generations to estimate uncertainty, which introduces substantial inference cost and often delays uncertainty assessment. In this paper, we investigate whether effective uncertainty estimation can be achieved with partial generation or even input-only information. Specifically, we first develop a unified framework that formulates uncertainty estimation as an early estimation problem over the autoregressive generation process of LLMs. This framework organises existing and proposed estimators by the information they observe, ranging from multi-generation to input-only prediction, and clarifies the performance-cost trade-off underlying different uncertainty estimation methods. Building on this view, we study two largely underexplored low-cost settings: estimating uncertainty with part of the generation, and predicting uncertainty from the input prompt. We propose Logit Magnitude, which uses top-M logit evidence to estimate uncertainty from an early-stopped generation prefix, and MetaUE, which distils generation-based uncertainty into a lightweight input-only estimator trained with uncertainty scores. Extensive experiments on general and domain-specific benchmarks show that Logit Magnitude achieves strong performance, and partial generations of LLMs are often sufficient for effective uncertainty estimation. MetaUE further provides a competitive input-only approximation in several settings. These findings suggest that effective uncertainty estimation requires less generation than commonly assumed, enabling unreliable responses to be identified earlier.

Zhikang Chen, Sen Cui, Deheng Ye et al.

Large Language Models (LLMs) have demonstrated strong reasoning capabilities through \emph{Chain-of-Thought} (CoT) prompting, which enables step-by-step intermediate reasoning. However, explicit CoT methods rely on discrete token-level reasoning processes that are prone to error propagation and limited by vocabulary expressiveness, often resulting in rigid and inconsistent reasoning trajectories. Recent research has explored implicit or continuous reasoning in latent spaces, allowing models to perform internal reasoning before generating explicit output. Although such approaches alleviate some limitations of discrete CoT, they generally lack explicit mechanisms to enforce consistency among reasoning steps, leading to divergent reasoning paths and unstable outcomes. To address this issue, we propose EBM-CoT, an Energy-Based Chain-of-Thought Calibration framework that refines latent thought representations through an energy-based model (EBM). Our method dynamically adjusts latent reasoning trajectories toward lower-energy, high-consistency regions in the embedding space, improving both reasoning accuracy and consistency without modifying the base language model. Extensive experiments across mathematical, commonsense, and symbolic reasoning benchmarks demonstrate that the proposed framework significantly enhances the consistency and efficiency of multi-step reasoning in LLMs.

Xianyuan Liu, Jiayang Zhang, Shuo Zhou et al.

Multimodal artificial intelligence (AI) integrates diverse types of data via machine learning to improve understanding, prediction, and decision-making across disciplines such as healthcare, science, and engineering. However, most multimodal AI advances focus on models for vision and language data, while their deployability remains a key challenge. We advocate a deployment-centric workflow that incorporates deployment constraints early to reduce the likelihood of undeployable solutions, complementing data-centric and model-centric approaches. We also emphasise deeper integration across multiple levels of multimodality and multidisciplinary collaboration to significantly broaden the research scope beyond vision and language. To facilitate this approach, we identify common multimodal-AI-specific challenges shared across disciplines and examine three real-world use cases: pandemic response, self-driving car design, and climate change adaptation, drawing expertise from healthcare, social science, engineering, science, sustainability, and finance. By fostering multidisciplinary dialogue and open research practices, our community can accelerate deployment-centric development for broad societal impact.

Chenqi Li, Timothy Denison, Tingting Zhu

Advancements in wearable sensor technologies and the digitization of medical records have contributed to the unprecedented ubiquity of biomedical time series data. Data-driven models have tremendous potential to assist clinical diagnosis and improve patient care by improving long-term monitoring capabilities, facilitating early disease detection and intervention, as well as promoting personalized healthcare delivery. However, accessing extensively labeled datasets to train data-hungry deep learning models encounters many barriers, such as long-tail distribution of rare diseases, cost of annotation, privacy and security concerns, data-sharing regulations, and ethical considerations. An emerging approach to overcome the scarcity of labeled data is to augment AI methods with human-like capabilities to leverage past experiences to learn new tasks with limited examples, called few-shot learning. This survey provides a comprehensive review and comparison of few-shot learning methods for biomedical time series applications. The clinical benefits and limitations of such methods are discussed in relation to traditional data-driven approaches. This paper aims to provide insights into the current landscape of few-shot learning for biomedical time series and its implications for future research and applications.

Ghadeer O. Ghosheh, Jin Li, Tingting Zhu

Electronic Health Records present a valuable modality for driving personalized medicine, where treatment is tailored to fit individual-level differences. For this purpose, many data-driven machine learning and statistical models rely on the wealth of longitudinal EHRs to study patients' physiological and treatment effects. However, longitudinal EHRs tend to be sparse and highly missing, where missingness could also be informative and reflect the underlying patient's health status. Therefore, the success of data-driven models for personalized medicine highly depends on how the EHR data is represented from physiological data, treatments, and the missing values in the data. To this end, we propose a novel deep-learning model that learns the underlying patient dynamics over time across multivariate data to generate personalized realistic values conditioning on an individual's demographic characteristics and treatments. Our proposed model, IGNITE (Individualized GeNeration of Imputations in Time-series Electronic health records), utilises a conditional dual-variational autoencoder augmented with dual-stage attention to generate missing values for an individual. In IGNITE, we further propose a novel individualized missingness mask (IMM), which helps our model generate values based on the individual's observed data and missingness patterns. We further extend the use of IGNITE from imputing missingness to a personalized data synthesizer, where it generates missing EHRs that were never observed prior or even generates new patients for various applications. We validate our model on three large publicly available datasets and show that IGNITE outperforms state-of-the-art approaches in missing data reconstruction and task prediction.

Mingcheng Zhu, Yu Liu, Zhiyao Luo et al.

Artificial Intelligence has revolutionised critical care for common conditions. Yet, rare conditions in the intensive care unit (ICU), including recognised rare diseases and low-prevalence conditions in the ICU, remain underserved due to data scarcity and intra-condition heterogeneity. To bridge such gaps, we developed KnowRare, a domain adaptation-based deep learning framework for predicting clinical outcomes for rare conditions in the ICU. KnowRare mitigates data scarcity by initially learning condition-agnostic representations from diverse electronic health records through self-supervised pre-training. It addresses intra-condition heterogeneity by selectively adapting knowledge from clinically similar conditions with a developed condition knowledge graph. Evaluated on two ICU datasets across five clinical prediction tasks (90-day mortality, 30-day readmission, ICU mortality, remaining length of stay, and phenotyping), KnowRare consistently outperformed existing state-of-the-art models. Additionally, KnowRare demonstrated superior predictive performance compared to established ICU scoring systems, including APACHE IV and IV-a. Case studies further demonstrated KnowRare's flexibility in adapting its parameters to accommodate dataset-specific and task-specific characteristics, its generalisation to common conditions under limited data scenarios, and its rationality in selecting source conditions. These findings highlight KnowRare's potential as a robust and practical solution for supporting clinical decision-making and improving care for rare conditions in the ICU.

Yu Liu, Weiyao Tao, Tong Xia et al.

Survival analysis, which estimates the probability of event occurrence over time from censored data, is fundamental in numerous real-world applications, particularly in high-stakes domains such as healthcare and risk assessment. Despite advances in numerous survival models, quantifying the uncertainty of predictions from these models remains underexplored and challenging. The lack of reliable uncertainty quantification limits the interpretability and trustworthiness of survival models, hindering their adoption in clinical decision-making and other sensitive applications. To bridge this gap, in this work, we introduce SurvUnc, a novel meta-model based framework for post-hoc uncertainty quantification for survival models. SurvUnc introduces an anchor-based learning strategy that integrates concordance knowledge into meta-model optimization, leveraging pairwise ranking performance to estimate uncertainty effectively. Notably, our framework is model-agnostic, ensuring compatibility with any survival model without requiring modifications to its architecture or access to its internal parameters. Especially, we design a comprehensive evaluation pipeline tailored to this critical yet overlooked problem. Through extensive experiments on four publicly available benchmarking datasets and five representative survival models, we demonstrate the superiority of SurvUnc across multiple evaluation scenarios, including selective prediction, misprediction detection, and out-of-domain detection. Our results highlight the effectiveness of SurvUnc in enhancing model interpretability and reliability, paving the way for more trustworthy survival predictions in real-world applications.

Munib Mesinovic, Soheila Molaei, Peter Watkinson et al.

Learning from longitudinal electronic health records is limited if it does not capture the temporal trajectories of the patient's state in a clinical setting. Graph models allow us to capture the hidden dependencies of the multivariate time-series when the graphs are constructed in a similar dynamic manner. Previous dynamic graph models require a pre-defined and/or static graph structure, which is unknown in most cases, or they only capture the spatial relations between the features. Furthermore in healthcare, the interpretability of the model is an essential requirement to build trust with clinicians. In addition to previously proposed attention mechanisms, there has not been an interpretable dynamic graph framework for data from multivariate electronic health records (EHRs). Here, we propose DynaGraph, an end-to-end interpretable contrastive graph model that learns the dynamics of multivariate time-series EHRs as part of optimisation. We validate our model in four real-world clinical datasets, ranging from primary care to secondary care settings with broad demographics, in challenging settings where tasks are imbalanced and multi-labelled. Compared to state-of-the-art models, DynaGraph achieves significant improvements in balanced accuracy and sensitivity over the nearest complex competitors in time-series or dynamic graph modelling across three ICU and one primary care datasets. Through a pseudo-attention approach to graph construction, our model also indicates the importance of clinical covariates over time, providing means for clinical validation.

Ghadeer O. Ghosheh, Jin Li, Tingting Zhu

With the widespread of machine learning models for healthcare applications, there is increased interest in building applications for personalized medicine. Despite the plethora of proposed research for personalized medicine, very few focus on representing missingness and learning from the missingness patterns in time-series Electronic Health Records (EHR) data. The lack of focus on missingness representation in an individualized way limits the full utilization of machine learning applications towards true personalization. In this brief communication, we highlight new insights into patterns of missingness with real-world examples and implications of missingness in EHRs. The insights in this work aim to bridge the gap between theoretical assumptions and practical observations in real-world EHRs. We hope this work will open new doors for exploring directions for better representation in predictive modelling for true personalization.

Zhikang Chen, Tingting Zhu

A world model is an AI system that simulates how an environment evolves under actions, enabling planning through imagined futures rather than reactive perception. Current world models, however, suffer from visual conflation: the mistaken assumption that high-fidelity video generation implies an understanding of physical and causal dynamics. We show that while modern models excel at predicting pixels, they frequently violate invariant constraints, fail under intervention, and break down in safety-critical decision-making. This survey argues that visual realism is an unreliable proxy for world understanding. Instead, effective world models must encode causal structure, respect domain-specific constraints, and remain stable over long horizons. We propose a reframing of world models as actionable simulators rather than visual engines, emphasizing structured 4D interfaces, constraint-aware dynamics, and closed-loop evaluation. Using medical decision-making as an epistemic stress test, where trial-and-error is impossible and errors are irreversible, we demonstrate that a world model's value is determined not by how realistic its rollouts appear, but by its ability to support counterfactual reasoning, intervention planning, and robust long-horizon foresight.

Chenqi Li, Yu Liu, Shuo Zhang et al.

Pretraining for electroencephalogram (EEG) foundation models has predominantly relied on self-supervised masked reconstruction, a paradigm largely adapted from and inspired by the success of vision and language foundation models. However, unlike images and text, EEG datasets are notoriously expensive to collect and characterized by low signal-to-noise ratio. These challenges introduce difficulties in scaling the EEG foundation models and capturing the underlying neural semantics through reconstruction. In this work, we ask the question: can we stand on the shoulders of well-established foundation models from well-represented modalities to bootstrap the pretraining of EEG foundation models? We first demonstrate that mainstream foundation models, such as those from vision and time series, transfer surprisingly well to EEG domain. To this end, we propose the Multi-Teacher Distillation Pretraining (MTDP) framework for pretraining EEG foundation models via a two-stage multi-teacher distillation. In the first stage, we introduce a learnable gating network to fuse representations from diverse teachers (e.g., DINOv3 and Chronos) via a masked latent denoising objective. In the second stage, we distill the fused representation into an EEG foundation model. Extensive evaluations across 9 downstream tasks and 12 datasets demonstrate that our MTDP-based EEG foundation model outperforms its self-supervised counterparts while requiring only 25% of the pretraining data.

Zecheng Pan, Zhikang Chen, Ding Li et al.

Merging models fine-tuned for different tasks into a single unified model has become an increasingly important direction for building versatile, efficient multi-task systems. Existing approaches predominantly rely on parameter interpolation in weight space, which we show introduces significant distribution shift in the feature space and undermines task-specific knowledge. In this paper, we propose OTMF (Optimal Transport-based Masked Fusion), a novel model merging framework rooted in optimal transport theory to address the distribution shift that arises from naive parameter interpolation. Instead of directly aggregating features or weights, OTMF aligns the semantic geometry of task-specific models by discovering common masks applied to task vectors through optimal transport plans. These masks selectively extract transferable and task-agnostic components while preserving the unique structural identities of each task. To ensure scalability in real-world settings, OTMF further supports a continual fusion paradigm that incrementally integrates each new task vector without revisiting previous ones, maintaining a bounded memory footprint and enabling efficient fusion across a growing number of tasks. We conduct comprehensive experiments on multiple vision and language benchmarks, and results show that OTMF achieves state-of-the-art performance in terms of both accuracy and efficiency. These findings highlight the practical and theoretical value of our approach to model merging.

Tianyi Chen, Mingcheng Zhu, Zhiyao Luo et al.

Electronic Health Records (EHRs) enable deep learning for clinical predictions, but the optimal method for representing patient data remains unclear due to inconsistent evaluation practices. We present the first systematic benchmark to compare EHR representation methods, including multivariate time-series, event streams, and textual event streams for LLMs. This benchmark standardises data curation and evaluation across two distinct clinical settings: the MIMIC-IV dataset for ICU tasks (mortality, phenotyping) and the EHRSHOT dataset for longitudinal care (30-day readmission, 1-year pancreatic cancer). For each paradigm, we evaluate appropriate modelling families--including Transformers, MLP, LSTMs and Retain for time-series, CLMBR and count-based models for event streams, 8-20B LLMs for textual streams--and analyse the impact of feature pruning based on data missingness. Our experiments reveal that event stream models consistently deliver the strongest performance. Pre-trained models like CLMBR are highly sample-efficient in few-shot settings, though simpler count-based models can be competitive given sufficient data. Furthermore, we find that feature selection strategies must be adapted to the clinical setting: pruning sparse features improves ICU predictions, while retaining them is critical for longitudinal tasks. Our results, enabled by a unified and reproducible pipeline, provide practical guidance for selecting EHR representations based on the clinical context and data regime.

Chenqi Li, Yu Liu, Timothy Denison et al.

Biosignals offer valuable insights into the physiological states of the human body. Although biosignal modalities differ in functionality, signal fidelity, sensor comfort, and cost, they are often intercorrelated, reflecting the holistic and interconnected nature of human physiology. This opens up the possibility of performing the same tasks using alternative biosignal modalities, thereby improving the accessibility, usability, and adaptability of health monitoring systems. However, the limited availability of large labeled datasets presents challenges for training models tailored to specific tasks and modalities of interest. Unsupervised cross-modal knowledge transfer offers a promising solution by leveraging knowledge from an existing modality to support model training for a new modality. Existing methods are typically based on knowledge distillation, which requires running a teacher model alongside student model training, resulting in high computational and memory overhead. This challenge is further exacerbated by the recent development of foundation models that demonstrate superior performance and generalization across tasks at the cost of large model sizes. To this end, we explore a new framework for unsupervised cross-modal knowledge transfer of biosignals by training a lightweight bridge network to align the intermediate representations and enable information flow between foundation models and across modalities. Specifically, we introduce an efficient strategy for selecting alignment positions where the bridge should be constructed, along with a flexible prototype network as the bridge architecture. Extensive experiments across multiple biosignal modalities, tasks, and datasets show that BioX-Bridge reduces the number of trainable parameters by 88--99\% while maintaining or even improving transfer performance compared to state-of-the-art methods.

Zi Cai, Yu Liu, Zhiyao Luo et al.

Digital healthcare systems have enabled the collection of mass healthcare data in electronic healthcare records (EHRs), allowing artificial intelligence solutions for various healthcare prediction tasks. However, existing studies often focus on isolated components of EHR data, limiting their predictive performance and interpretability. To address this gap, we propose ProtoEHR, an interpretable hierarchical prototype learning framework that fully exploits the rich, multi-level structure of EHR data to enhance healthcare predictions. More specifically, ProtoEHR models relationships within and across three hierarchical levels of EHRs: medical codes, hospital visits, and patients. We first leverage large language models to extract semantic relationships among medical codes and construct a medical knowledge graph as the knowledge source. Building on this, we design a hierarchical representation learning framework that captures contextualized representations across three levels, while incorporating prototype information within each level to capture intrinsic similarities and improve generalization. To perform a comprehensive assessment, we evaluate ProtoEHR in two public datasets on five clinically significant tasks, including prediction of mortality, prediction of readmission, prediction of length of stay, drug recommendation, and prediction of phenotype. The results demonstrate the ability of ProtoEHR to make accurate, robust, and interpretable predictions compared to baselines in the literature. Furthermore, ProtoEHR offers interpretable insights on code, visit, and patient levels to aid in healthcare prediction.

Rui Feng, Zhiyao Luo, Wei Wang et al.

Automatic assessment of cognitive impairment from spontaneous speech offers a promising, non-invasive avenue for early cognitive screening. However, current approaches often lack generalizability when deployed across different languages and clinical settings, limiting their practical utility. In this study, we propose CogBench, the first benchmark designed to evaluate the cross-lingual and cross-site generalizability of large language models (LLMs) for speech-based cognitive impairment assessment. Using a unified multimodal pipeline, we evaluate model performance on three speech datasets spanning English and Mandarin: ADReSSo, NCMMSC2021-AD, and a newly collected test set, CIR-E. Our results show that conventional deep learning models degrade substantially when transferred across domains. In contrast, LLMs equipped with chain-of-thought prompting demonstrate better adaptability, though their performance remains sensitive to prompt design. Furthermore, we explore lightweight fine-tuning of LLMs via Low-Rank Adaptation (LoRA), which significantly improves generalization in target domains. These findings offer a critical step toward building clinically useful and linguistically robust speech-based cognitive assessment tools.

Jianqiao Chen, Tingting Zhu, Huishi Song et al.

Discretization of semantic features enables interoperability between semantic and digital communication systems, showing significant potential for practical applications. The fundamental difficulty in digitizing semantic features stems from the need to preserve continuity and context in inherently analog representations during their compression into discrete symbols while ensuring robustness to channel degradation. In this paper, we propose a vector quantized (VQ)-enabled digital semantic communication system with channel adaptive image transmission, named VQ-DeepISC. Guided by deep joint source-channel coding (DJSCC), we first design a Swin Transformer backbone for hierarchical semantic feature extraction, followed by VQ modules projecting features into discrete latent spaces. Consequently, it enables efficient index-based transmission instead of raw feature transmission. To further optimize this process, we develop an attention mechanism-driven channel adaptation module to dynamically optimize index transmission. Secondly, to counteract codebook collapse during training process, we impose a distributional regularization by minimizing the Kullback-Leibler divergence (KLD) between codeword usage frequencies and a uniform prior. Meanwhile, exponential moving average (EMA) is employed to stabilize training and ensure balanced feature coverage during codebook updates. Finally, digital communication is implemented using quadrature phase shift keying (QPSK) modulation alongside orthogonal frequency division multiplexing (OFDM), adhering to the IEEE 802.11a standard. Experimental results demonstrate superior reconstruction fidelity of the proposed system over benchmark methods.

Yangyang Xu, Bangzhen Liu, Wenqi Shao et al.

Decoding stimulus images from fMRI signals has advanced with pre-trained generative models. However, existing methods struggle with cross-subject mappings due to cognitive variability and subject-specific differences. This challenge arises from sequential errors, where unidirectional mappings generate partially inaccurate representations that, when fed into diffusion models, accumulate errors and degrade reconstruction fidelity. To address this, we propose the Bidirectional Autoencoder Intertwining framework for accurate decoded representation prediction. Our approach unifies multiple subjects through a Subject Bias Modulation Module while leveraging bidirectional mapping to better capture data distributions for precise representation prediction. To further enhance fidelity when decoding representations into stimulus images, we introduce a Semantic Refinement Module to improve semantic representations and a Visual Coherence Module to mitigate the effects of inaccurate visual representations. Integrated with ControlNet and Stable Diffusion, our method outperforms state-of-the-art approaches on benchmark datasets in both qualitative and quantitative evaluations. Moreover, our framework exhibits strong adaptability to new subjects with minimal training samples.

Tingting Zhu, Tingyang Chen, Yinghui Wu et al.

Ensuring the trustworthiness of graph neural networks (GNNs) as black-box models requires effective explanation methods. Existing GNN explanations typically apply input perturbations to identify subgraphs that are responsible for the occurrence of the final output of GNNs. However, such approaches lack finer-grained, layer-wise analysis of how intermediate representations contribute to the final result, capabilities that are crucial for model diagnosis and architecture optimization. This paper introduces SliceGX, a novel GNN explanation approach that generates explanations at specific GNN layers in a progressive manner. Given a GNN M, a set of selected intermediate layers, and a target layer, SliceGX automatically segments M into layer blocks ("model slice") and discovers high-quality explanatory subgraphs in each layer block that clarifies the occurrence of output of M at the targeted layer. Although finding such layer-wise explanations is computationally challenging, we develop efficient algorithms and optimization techniques that incrementally generate and maintain these subgraphs with provable approximation guarantees. Additionally, SliceGX offers a SPARQL-like query interface, providing declarative access and search capacities for the generated explanations. Through experiments on large real-world graphs and representative GNN architectures, we verify the effectiveness and efficiency of SliceGX, and illustrate its practical utility in supporting model debugging.

Chenqi Li, Boyan Gao, Gabriel Jones et al.

Deep learning models have demonstrated exceptional performance in a variety of real-world applications. These successes are often attributed to strong base models that can generalize to novel tasks with limited supporting data while keeping prior knowledge intact. However, these impressive results are based on the availability of a large amount of high-quality data, which is often lacking in specialized biomedical applications. In such fields, models are usually developed with limited data that arrive incrementally with novel categories. This requires the model to adapt to new information while preserving existing knowledge. Few-Shot Class-Incremental Learning (FSCIL) methods offer a promising approach to addressing these challenges, but they also depend on strong base models that face the same aforementioned limitations. To overcome these constraints, we propose AnchorInv following the straightforward and efficient buffer-replay strategy. Instead of selecting and storing raw data, AnchorInv generates synthetic samples guided by anchor points in the feature space. This approach protects privacy and regularizes the model for adaptation. When evaluated on three public physiological time series datasets, AnchorInv exhibits efficient knowledge forgetting prevention and improved adaptation to novel classes, surpassing state-of-the-art baselines.

Ghadeer O. Ghosheh, Moritz Gögl, Tingting Zhu

The burden of diseases is rising worldwide, with unequal treatment efficacy for patient populations that are underrepresented in clinical trials. Healthcare, however, is driven by the average population effect of medical treatments and, therefore, operates in a "one-size-fits-all" approach, not necessarily what best fits each patient. These facts suggest a pressing need for methodologies to study individualized treatment effects (ITE) to drive personalized treatment. Despite the increased interest in machine-learning-driven ITE estimation models, the vast majority focus on tabular data with limited review and understanding of methodologies proposed for time-series electronic health records (EHRs). To this end, this work provides an overview of ITE works for time-series data and insights into future research. The work summarizes the latest work in the literature and reviews it in light of theoretical assumptions, types of treatment settings, and computational frameworks. Furthermore, this work discusses challenges and future research directions for ITEs in a time-series setting. We hope this work opens new directions and serves as a resource for understanding one of the exciting yet under-studied research areas.

Munib Mesinovic, Peter Watkinson, Tingting Zhu

Heart attack remain one of the greatest contributors to mortality in the United States and globally. Patients admitted to the intensive care unit (ICU) with diagnosed heart attack (myocardial infarction or MI) are at higher risk of death. In this study, we use two retrospective cohorts extracted from the eICU and MIMIC-IV databases, to develop a novel pseudo-dynamic machine learning framework for mortality prediction in the ICU with interpretability and clinical risk analysis. The method provides accurate prediction for ICU patients up to 24 hours before the event and provide time-resolved interpretability results. The performance of the framework relying on extreme gradient boosting was evaluated on a held-out test set from eICU, and externally validated on the MIMIC-IV cohort using the most important features identified by time-resolved Shapley values achieving AUCs of 91.0 (balanced accuracy of 82.3) for 6-hour prediction of mortality respectively. We show that our framework successfully leverages time-series physiological measurements by translating them into stacked static prediction problems to be robustly predictive through time in the ICU stay and can offer clinical insight from time-resolved interpretability

Yujiang Wang, Anshul Thakur, Mingzhi Dong et al.

The prevalence of artificial intelligence (AI) has envisioned an era of healthcare democratisation that promises every stakeholder a new and better way of life. However, the advancement of clinical AI research is significantly hurdled by the dearth of data democratisation in healthcare. To truly democratise data for AI studies, challenges are two-fold: 1. the sensitive information in clinical data should be anonymised appropriately, and 2. AI-oriented clinical knowledge should flow freely across organisations. This paper considers a recent deep-learning advent, dataset condensation (DC), as a stone that kills two birds in democratising healthcare data. The condensed data after DC, which can be viewed as statistical metadata, abstracts original clinical records and irreversibly conceals sensitive information at individual levels; nevertheless, it still preserves adequate knowledge for learning deep neural networks (DNNs). More favourably, the compressed volumes and the accelerated model learnings of condensed data portray a more efficient clinical knowledge sharing and flowing system, as necessitated by data democratisation. We underline DC's prospects for democratising clinical data, specifically electrical healthcare records (EHRs), for AI research through experimental results and analysis across three healthcare datasets of varying data types.

Anshul Thakur, Tingting Zhu, Vinayak Abrol et al.

The lack of data democratization and information leakage from trained models hinder the development and acceptance of robust deep learning-based healthcare solutions. This paper argues that irreversible data encoding can provide an effective solution to achieve data democratization without violating the privacy constraints imposed on healthcare data and clinical models. An ideal encoding framework transforms the data into a new space where it is imperceptible to a manual or computational inspection. However, encoded data should preserve the semantics of the original data such that deep learning models can be trained effectively. This paper hypothesizes the characteristics of the desired encoding framework and then exploits random projections and random quantum encoding to realize this framework for dense and longitudinal or time-series data. Experimental evaluation highlights that models trained on encoded time-series data effectively uphold the information bottleneck principle and hence, exhibit lesser information leakage from trained models.

Alex Youssef, Michael Pencina, Anshul Thakur et al.

External validation is often recommended to ensure the generalizability of ML models. However, it neither guarantees generalizability nor equates to a model's clinical usefulness (the ultimate goal of any clinical decision-support tool). External validation is misaligned with current healthcare ML needs. First, patient data changes across time, geography, and facilities. These changes create significant volatility in the performance of a single fixed model (especially for deep learning models, which dominate clinical ML). Second, newer ML techniques, current market forces, and updated regulatory frameworks are enabling frequent updating and monitoring of individual deployed model instances. We submit that external validation is insufficient to establish ML models' safety or utility. Proposals to fix the external validation paradigm do not go far enough. Continued reliance on it as the ultimate test is likely to lead us astray. We propose the MLOps-inspired paradigm of recurring local validation as an alternative that ensures the validity of models while protecting against performance-disruptive data variability. This paradigm relies on site-specific reliability tests before every deployment, followed by regular and recurrent checks throughout the life cycle of the deployed algorithm. Initial and recurrent reliability tests protect against performance-disruptive distribution shifts, and concept drifts that jeopardize patient safety.

Jin Li, Benjamin J. Cairns, Jingsong Li et al.

The recent availability of electronic health records (EHRs) have provided enormous opportunities to develop artificial intelligence (AI) algorithms. However, patient privacy has become a major concern that limits data sharing across hospital settings and subsequently hinders the advances in AI. Synthetic data, which benefits from the development and proliferation of generative models, has served as a promising substitute for real patient EHR data. However, the current generative models are limited as they only generate single type of clinical data for a synthetic patient, i.e., either continuous-valued or discrete-valued. To mimic the nature of clinical decision-making which encompasses various data types/sources, in this study, we propose a generative adversarial network (GAN) entitled EHR-M-GAN which simultaneously synthesizes mixed-type timeseries EHR data. EHR-M-GAN is capable of capturing the multidimensional, heterogeneous, and correlated temporal dynamics in patient trajectories. We have validated EHR-M-GAN on three publicly-available intensive care unit databases with records from a total of 141,488 unique patients, and performed privacy risk evaluation of the proposed model. EHR-M-GAN has demonstrated its superiority over state-of-the-art benchmarks for synthesizing clinical timeseries with high fidelity, while addressing the limitations regarding data types and dimensionality in the current generative models. Notably, prediction models for outcomes of intensive care performed significantly better when training data was augmented with the addition of EHR-M-GAN-generated timeseries. EHR-M-GAN may have use in developing AI algorithms in resource-limited settings, lowering the barrier for data acquisition while preserving patient privacy.

Rasheed el-Bouri, Tingting Zhu, David A. Clifton

Given the abundance and ease of access of personal data today, individual privacy has become of paramount importance, particularly in the healthcare domain. In this work, we aim to utilise patient data extracted from multiple hospital data centres to train a machine learning model without sacrificing patient privacy. We develop a scheduling algorithm in conjunction with a student-teacher algorithm that is deployed in a federated manner. This allows a central model to learn from batches of data at each federal node. The teacher acts between data centres to update the main task (student) algorithm using the data that is stored in the various data centres. We show that the scheduler, trained using meta-gradients, can effectively organise training and as a result train a machine learning model on a diverse dataset without needing explicit access to the patient data. We achieve state-of-the-art performance and show how our method overcomes some of the problems faced in the federated learning such as node poisoning. We further show how the scheduler can be used as a mechanism for transfer learning, allowing different teachers to work together in training a student for state-of-the-art performance.

Girmaw Abebe Tadesse, Hamza Javed, Yong Liu et al.

Myocardial Infarction (MI) has the highest mortality of all cardiovascular diseases (CVDs). Detection of MI and information regarding its occurrence-time in particular, would enable timely interventions that may improve patient outcomes, thereby reducing the global rise in CVD deaths. Electrocardiogram (ECG) recordings are currently used to screen MI patients. However, manual inspection of ECGs is time-consuming and prone to subjective bias. Machine learning methods have been adopted for automated ECG diagnosis, but most approaches require extraction of ECG beats or consider leads independently of one another. We propose an end-to-end deep learning approach, DeepMI, to classify MI from normal cases as well as identifying the time-occurrence of MI (defined as acute, recent and old), using a collection of fusion strategies on 12 ECG leads at data-, feature-, and decision-level. In order to minimise computational overhead, we employ transfer learning using existing computer vision networks. Moreover, we use recurrent neural networks to encode the longitudinal information inherent in ECGs. We validated DeepMI on a dataset collected from 17,381 patients, in which over 323,000 samples were extracted per ECG lead. We were able to classify normal cases as well as acute, recent and old onset cases of MI, with AUROCs of 96.7%, 82.9%, 68.6% and 73.8%, respectively. We have demonstrated a multi-lead fusion approach to detect the presence and occurrence-time of MI. Our end-to-end framework provides flexibility for different levels of multi-lead ECG fusion and performs feature extraction via transfer learning.

Dani Kiyasseh, Tingting Zhu, David Clifton

Cardiac signals, such as the electrocardiogram, convey a significant amount of information about the health status of a patient which is typically summarized by a clinician in the form of a clinical report, a cumbersome process that is prone to errors. To streamline this routine process, we propose a deep neural network capable of captioning cardiac signals; it receives a cardiac signal as input and generates a clinical report as output. We extend this further to generate multilingual reports. To that end, we create and make publicly available a multilingual clinical report dataset. In the absence of sufficient labelled data, deep neural networks can benefit from a warm-start, or pre-training, procedure in which parameters are first learned in an arbitrary task. We propose such a task in the form of discriminative multilingual pre-training where tokens from clinical reports are randomly replaced with those from other languages and the network is tasked with predicting the language of all tokens. We show that our method performs on par with state-of-the-art pre-training methods such as MLM, ELECTRA, and MARGE, while simultaneously generating diverse and plausible clinical reports. We also demonstrate that multilingual models can outperform their monolingual counterparts, informally terming this beneficial phenomenon as the blessing of multilinguality.

Dani Kiyasseh, Tingting Zhu, David A. Clifton

The process of manually searching for relevant instances in, and extracting information from, clinical databases underpin a multitude of clinical tasks. Such tasks include disease diagnosis, clinical trial recruitment, and continuing medical education. This manual search-and-extract process, however, has been hampered by the growth of large-scale clinical databases and the increased prevalence of unlabelled instances. To address this challenge, we propose a supervised contrastive learning framework, CROCS, where representations of cardiac signals associated with a set of patient-specific attributes (e.g., disease class, sex, age) are attracted to learnable embeddings entitled clinical prototypes. We exploit such prototypes for both the clustering and retrieval of unlabelled cardiac signals based on multiple patient attributes. We show that CROCS outperforms the state-of-the-art method, DTC, when clustering and also retrieves relevant cardiac signals from a large database. We also show that clinical prototypes adopt a semantically meaningful arrangement based on patient attributes and thus confer a high degree of interpretability.

Dani Kiyasseh, Tingting Zhu, David A. Clifton

Many clinical deep learning algorithms are population-based and difficult to interpret. Such properties limit their clinical utility as population-based findings may not generalize to individual patients and physicians are reluctant to incorporate opaque models into their clinical workflow. To overcome these obstacles, we propose to learn patient-specific embeddings, entitled patient cardiac prototypes (PCPs), that efficiently summarize the cardiac state of the patient. To do so, we attract representations of multiple cardiac signals from the same patient to the corresponding PCP via supervised contrastive learning. We show that the utility of PCPs is multifold. First, they allow for the discovery of similar patients both within and across datasets. Second, such similarity can be leveraged in conjunction with a hypernetwork to generate patient-specific parameters, and in turn, patient-specific diagnoses. Third, we find that PCPs act as a compact substitute for the original dataset, allowing for dataset distillation.

Henrique Aguiar, Mauro Santos, Peter Watkinson et al.

Recent years have seen an increased focus into the tasks of predicting hospital inpatient risk of deterioration and trajectory evolution due to the availability of electronic patient data. A common approach to these problems involves clustering patients time-series information such as vital sign observations) to determine dissimilar subgroups of the patient population. Most clustering methods assume time-invariance of vital-signs and are unable to provide interpretability in clusters that is clinically relevant, for instance, event or outcome information. In this work, we evaluate three different clustering models on a large hospital dataset of vital-sign observations from patients suffering from Chronic Obstructive Pulmonary Disease. We further propose novel modifications to deal with unevenly sampled time-series data and unbalanced class distribution to improve phenotype separation. Lastly, we discuss further avenues of investigation for models to learn patient subgroups with distinct behaviour and phenotype.

Rasheed el-Bouri, David Eyre, Peter Watkinson et al.

Accurate and reliable prediction of hospital admission location is important due to resource-constraints and space availability in a clinical setting, particularly when dealing with patients who come from the emergency department. In this work we propose a student-teacher network via reinforcement learning to deal with this specific problem. A representation of the weights of the student network is treated as the state and is fed as an input to the teacher network. The teacher network's action is to select the most appropriate batch of data to train the student network on from a training set sorted according to entropy. By validating on three datasets, not only do we show that our approach outperforms state-of-the-art methods on tabular data and performs competitively on image recognition, but also that novel curricula are learned by the teacher network. We demonstrate experimentally that the teacher network can actively learn about the student network and guide it to achieve better performance than if trained alone.

Dani Kiyasseh, Tingting Zhu, David A. Clifton

The healthcare industry generates troves of unlabelled physiological data. This data can be exploited via contrastive learning, a self-supervised pre-training method that encourages representations of instances to be similar to one another. We propose a family of contrastive learning methods, CLOCS, that encourages representations across space, time, \textit{and} patients to be similar to one another. We show that CLOCS consistently outperforms the state-of-the-art methods, BYOL and SimCLR, when performing a linear evaluation of, and fine-tuning on, downstream tasks. We also show that CLOCS achieves strong generalization performance with only 25\% of labelled training data. Furthermore, our training procedure naturally generates patient-specific representations that can be used to quantify patient-similarity.

Dani Kiyasseh, Tingting Zhu, David A. Clifton

Large sets of unlabelled data within the healthcare domain remain underutilized. Active learning offers a way to exploit these datasets by iteratively requesting an oracle (e.g. medical professional) to label instances. This process, which can be costly and time-consuming is overly-dependent upon an oracle. To alleviate this burden, we propose SoQal, a questioning strategy that dynamically determines when a label should be requested from an oracle. We perform experiments on five publically-available datasets and illustrate SoQal's superiority relative to baseline approaches, including its ability to reduce oracle label requests by up to 35%. SoQal also performs competitively in the presence of label noise: a scenario that simulates clinicians' uncertain diagnoses when faced with difficult classification tasks.