医学 AI

Weakly supervised segmentation enables model training from plane-level labels. Existing methods often rely on 2D encoders, neglecting the volumetric nature of medical data. We propose TranSamba, a hybrid Transformer-Mamba architecture designed to capture 3D context via cross-plane modeling. TranSamba augments a Vision Transformer backbone with Cross-Plane Mamba blocks, leveraging linear-time modeling for efficient information exchange across neighboring planes. This exchange improves in-plane self-attention and subsequent attention maps for object localization. TranSamba maintains linear time complexity and constant space complexity with respect to the input volume depth. Extensive experiments on three datasets covering diverse modalities and pathologies show that TranSamba achieves state-of-the-art performance, demonstrating the generalizable efficacy of cross-plane modeling. Code is available at: https://github.com/YihengLyu/TranSamba.

Volumetric ultrasound imaging faces a fundamental trade-off among image quality, frame rate, and hardware complexity. This study introduces three-dimensional Null Subtraction Imaging (3D NSI), a nonlinear beamforming framework that addresses this trade-off by combining computationally efficient null-subtraction process with multiplexing-aware sparse aperture designs on matrix arrays. We evaluate three apodization configurations: a fully addressed circular aperture and two Fermat's spiral sparse apertures. To overcome channel-sharing constraints common in matrix arrays multiplexed with low-channel-count ultrasound systems, we propose a spiral "no-reuse" apodization that enforces non-overlapping element sets across transmit-receive events. This design resolves multiplexing conflicts and enables up to a 16-fold increase in acquisition volume rate using only 240 active elements on a 1024-element probe. In computer simulations and tissue-mimicking phantom experiments, 3D NSI achieved an average improvement of 36% in azimuthal and elevational resolutions, along with an approximately 20% higher contrast ratio, compared to the conventional Delay-and-Sum (DAS) beamformer under matched transmit/receive configurations. When implemented with the spiral no-reuse aperture, the 3D NSI framework achieved over 1000 volumes per second with a computational load less than three times that of DAS, making it a practical solution for real-time 4D imaging.

In positron emission tomography (PET), it is indispensable to perform attenuation correction in order to obtain the quantitatively accurate activity map (tracer distribution) in the body. Generally, this is carried out based on the estimated attenuation map obtained from computed tomography or magnetic resonance imaging. However, except for errors in the attenuation correction factors obtained, the additional scan not only brings in new radiation doses and/or increases the scanning time but also leads to severe misalignment induced by various motions during and between the two sequential scans. To address these issues, based on maximum likelihood estimation, we propose a new mathematical model for simultaneously reconstructing the activity and attenuation sinogram from the time-of-flight (TOF)-PET emission data only. Particularly, we make full use of the exclusively exponential form for the attenuation correction factors, and consider the constraint of a total amount of the activity in some mask region in the proposed model. Furthermore, we prove its well-posedness, including the existence, uniqueness and stability of the solution. We propose an alternating update algorithm to solve the model, and also analyze its convergence. Finally, numerical experiments with various TOF-PET emission data demonstrate that the proposed method is of numerical convergence and robust to noise, and outperforms some state-of-the-art methods in terms of accuracy and efficiency, and has the capability of autonomous attenuation correction.

Limited-Angle Computed Tomography (LACT) is a challenging inverse problem where missing angular projections lead to incomplete sinograms and severe artifacts in the reconstructed images. While recent learning-based methods have demonstrated effectiveness, most of them assume ideal, noise-free measurements and fail to address the impact of measurement noise. To overcome this limitation, we treat LACT as a sinogram inpainting task and propose a diffusion-based framework that completes missing angular views using a Mean-Reverting Stochastic Differential Equation (MR-SDE) formulation. To improve robustness under realistic noise, we propose RNSD$^+$, a novel noise-aware rectification mechanism that explicitly models inference-time uncertainty, enabling reliable and robust reconstruction. Extensive experiments demonstrate that our method consistently surpasses baseline models in data consistency and perceptual quality, and generalizes well across varying noise intensity and acquisition scenarios.

Objective: Deep Learning has shown promise in accelerating MRI by reconstructing high-quality images from under-sampled data. While recent work has leveraged multi-contrast information to improve reconstruction performance, these methods rely on supervised learning, which requires fully sampled k-space for training. One method, self-supervised learning via data undersampling (SSDU), enables direct training on under-sampled k-space by partitioning it into two sets, with a network mapping between the two. In this work, we improve MRI self-supervised MRI reconstruction with two modifications. Methods: We propose a multi-contrast self-supervised learning framework that jointly trains on multiple under-sampled contrasts without requiring fully sampled k-space data as a reference. Moreover, we learn an optimal self-supervised data partitioning for each contrast in an end-to-end manner, further enhancing reconstruction quality. Specifically, we learn an optimal partitioning probability distribution, which is sampled to generate a mask for partitioning. Results: Experiments on two publicly available multi-contrast MRI datasets demonstrate the improved reconstruction quality of our proposed self-supervised multi-contrast learned partitioning method compared to the current single-contrast self-supervised learning methods. We also demonstrate that learning the partitioning of k-space data further enhances the fidelity of reconstructions. Conclusion: Multi-contrast reconstruction combined with learned partitioning improves reconstruction fidelity over single-contrast self-supervised MRI reconstructions. Significance: Our method can facilitate higher image fidelity and/or accelerated MRI protocol times compared to previous self-supervised methods, and without requiring fully sampled k-space for training.

Recent advances in deep learning have significantly accelerated cardiac imaging workflows, from segmentation to the generation of meshes for computational modelling. Nevertheless, analysis of 3D echocardiograms presents unique challenges due to their low contrast-to-noise ratio, conical field of view, and susceptibility to acoustic shadowing. Here, we present an efficient and practical network tailored for 3D echocardiograms. Our method consists of a two-stage network that combines convolutional neural networks, graph convolutional networks, and transformers, to create accurate time-varying 3D meshes of the left ventricle that are topologically consistent and temporally coherent throughout the cardiac cycle. Our model achieved superior mesh reconstruction accuracy compared to current state-of-the-art methods on a held-out test dataset of 100 3D echo images, with a Dice coefficient of 0.87 +/- 0.05 (cavity) and 0.75 +/- 0.07 (myocardium), and mean +/- SD surface distances of 3.3 +/- 0.6 mm (endocardium) and 3.5 +/- 0.5 mm (epicardium), against reference segmentations derived from cardiac magnetic resonance imaging. The reconstructed mesh enables automated calculation of routine clinical indices, such as volume, mass, and strain, and enables advanced applications with biophysical digital twins. Source code is openly shared at https://github.com/EdwardFerdian/ghost-cat.

Zero-shot anomaly detection (ZSAD) is attractive for medical imaging because clinical systems must handle heterogeneous acquisition protocols, changing patient populations, and pathologies for which annotated training data may be unavailable. Most existing zero-shot anomaly detection methods are designed for 2D images, and their direct extension to 3D medical volumes is limited by the scarcity of large-scale volumetric foundation models or by the difficulty of utilizing volumetric context. We propose CS3F, a training-free batch-based framework for ZSAD in 3D medical images using 2D foundation models. Each volume is decomposed along multiple anatomical axes and encoded slice-wise by a 2D vision transformer. These are then converted into localized volumetric tokens by pooling neighboring slice features. Anomaly scores are obtained from cross-subject mutual similarity: tokens that lack close analogues in other subjects are assigned higher anomaly scores. To reduce the attenuation of focal lesion signals caused by depth pooling, we introduce a coarse-to-fine tokenization strategy that enables fine-resolution volumetric scoring without exhaustive matching. CS3F is evaluated on brain MRI across metastases, glioma, and stroke, as well as validated on lung CT to test generalizability beyond atlas-aligned brain MRI. The results show that frozen 2D foundation models can support anomaly localization in 3D medical images, and that the benefit of fine tokenization depends strongly on lesion contrast and imaging modality.

Learning unsupervised representations of medical imaging cohorts can reveal clinically meaningful prototypes without expert labels, which are often noisy and fail to capture true pathological heterogeneity. However, existing deep latent-variable models estimate Gaussian mixture priors via Euclidean averaging, producing prototypes that drift off the curved data manifold and degenerate as the number of sub-populations grows. We propose a manifold-anchored variational framework built on a geometry-aware Expectation-Maximization (EM) algorithm, whose M-step selects each sub-population prototype as the graph medoid with the highest diffusion centrality on a heat-kernel-weighted latent graph, ensuring that every prototype remains on-manifold. A Dirichlet energy regularizer enforces geometric smoothness of the latent space, and a per-sub-population uncertainty score enables label-free quality assessment. \rev{The manifold-anchored EM is a general-purpose geometric tool that extends standard EM and applies readily to other latent-variable models beyond this setting.} On cardiac scar and brain MRI benchmarks, our framework attains the highest accuracy among all compared methods, produces the sharpest prototypes reported to date, and remains stable at large sub-population counts where all baselines degenerate.

Pathological images are inherently multi-scale, requiring pathologists to integrate evidence from global tissue architecture at low magnification to cellular morphology at higher magnification for accurate diagnosis. While existing pathological datasets for vision-language model (VLM) include various scales, they often lack an explicit cross-scale reasoning objective. This limitation prevents VLMs from capturing essential cross-scale representations and learning evidence-based reasoning. To bridge this gap, we introduce the first cross-scale training and evaluation paradigm that formulates pathology interpretation as multi-magnification reasoning. However, creating such a task reveals a critical challenge: multi-image visual question answering (VQA) is prone to text-only shortcuts, which allow models to guess answers using magnification-dependent artifacts rather than visual evidence. To address this, we propose a leakage-aware curation pipeline that combines adversarial text-only screening with constraint-guided question design. Using this pipeline, we construct Scale-VQA, a high-quality benchmark with 4,685 multiple-choice questions grounded in 2,537 pathology images across multiple magnification levels. Finally, we present ScaleReasoner-R1, a model trained via reinforcement learning to optimize performance on the cross-scale VQA task. ScaleReasoner-R1 achieves state-of-the-art performance on our cross-scale reasoning benchmark and generalizes to SOTA performance on established single-scale benchmarks. Findings suggest that even the limited cross-scale supervision can significantly improve pathological understanding. The code and demos will be open-sourced.

In-silico trials of medical devices require the generation of virtual populations of anatomies. In cardiovascular applications, virtual anatomy is typically represented as a 3D+t mesh sampled from a generative model. However, most existing mesh generators focus on static anatomy, while sequence models often lack explicit periodicity. To this end, we propose 4D F-MeshLDM, a conditional generative framework comprising a convolutional mesh VAE to encode meshes, a structural latent space that parameterises motion using a truncated Fourier series, and a diffusion prior that learns the latent distribution over Fourier coefficient tokens. By conditioning the diffusion process on clinical covariates via affine modulation, we enable controllable synthesis. Sampling tokens and performing inverse Fourier synthesis yield cycle-consistent latent trajectories, which can be decoded into 3D+t cardiac mesh sequences. Experiments on 5,000 UK Biobank subjects demonstrate that 4D F-MeshLDM outperforms state-of-the-art baselines in anatomical fidelity and achieves near-zero cycle closure error. Furthermore, the generated cohorts accurately preserve clinical functional indices, highlighting the potential of our framework for reliable in-silico cardiac trials.

Avoiding the risk of undefined categorical labels using nearest neighbor interpolation overlooks the risk of exacerbating pixel level annotation errors in augmented training data. Additionally, the inherent low pass filtering effects of interpolation algorithms exacerbate the risk of degrading high frequency structural details within annotated regions of interest. To avoid these risks, the author modified convolutional neural networks data transformation functions by incorporating a modified geometric transformation function, removing reliance on nearest neighbor interpolation, and integrating a mean-based class filtering mechanism to handle undefined categorical labels with alternative interpolation algorithms. The author also implemented an offline data augmentation pipeline to generate interpolation specific augmented training data, enabling quantitative assessment of interpolation specific low pass filtering effects on augmented training data. Experimental evaluation on three medical image segmentation datasets and the XBAT+ datasets demonstrated performance gains across multiple quantitative metrics.

Clinician-centered evaluation is critical for validating medical AI systems, especially in ultrasound imaging where quantitative metrics do not always capture clinical usability. Existing medical image platforms primarily focus on dataset labeling. They lack integrated support for blinded model comparison and reproducible evaluation workflows. We present a clinician-centered pipeline for remote annotation and evaluation in ultrasound AI studies. The proposed pipeline uses a centralized server and lightweight browser interfaces to enable clinicians to perform annotation, blinded ranking, and review without local dataset downloads. The pipeline also supports multi-rater participation, centralized result aggregation, and automated statistical analysis. We validate the pipeline in a fetal ultrasound segmentation study with six raters spanning expert, generalist, and non-expert experience levels. The system automatically generated Spearman correlation, Kendall's $τ$, and top-1 selection statistics. Results indicated moderate to strong agreement across experts and other groups. The blinded evaluation results showed a tendency for later active learning models to be preferred. These outcomes suggest that the pipeline can support clinician-centered annotation and reproducible human-\ac{AI} evaluation studies in ultrasound imaging. The proposed pipeline is available on \href{https://github.com/13204942/SonoRate}{GitHub}.

Multimodal neuroimaging, integrating functional connectivity from fMRI and structural connectivity from DTI, enables non-invasive analysis of brain networks using graph neural networks. However, demographic factors such as age and sex systematically confound the relationship between brain connectivity and clinical outcomes, causing GNNs to exploit spurious shortcuts rather than learning causally invariant representations. While recent causal GNN methods introduce causality at the graph-modeling level, their causal mechanisms remain domain-agnostic without accounting for the real-world confounders inherent in clinical neuroimaging data. Moreover, brain networks are constructed from atlas-based parcellations where each region exhibits distinct sensitivity to demographic factors, necessitating region-aware adjustment. We propose Artemis, a region-level causal framework that bridges this gap with causal intervention at each brain region independently by learning region-specific confounder representations with lightweight parameters. Our adjustment comprehensively utilized the multimodal functional and structural features for graph reasoning as a plug-in module compatible with arbitrary GNN backbones. Experiments on three benchmarks, ADNI for disease diagnosis, OASIS for dementia staging, and HCP for sex classification, demonstrate consistent improvements over representative GNN-based baselines. Multiple supporting experiments further demonstrate statistical significance and neuroscientific interpretability.

Artificial intelligence has driven rapid progress in medical imaging research, producing increasingly sophisticated algorithms and steady improvements on benchmark tasks. However, this algorithm-centric trajectory has also revealed a growing imbalance: while computational methods advance rapidly, the conceptual foundations that define imaging tasks, evaluation metrics, and clinical meaning sometimes remain underexamined. In this Perspective, we distinguish algorithmic innovation, which focuses on improving computational implementations and performance within a fixed problem definition, from conceptual innovation, which reframes what problems are posed, how success is measured, and why an approach is clinically relevant. We argue that prevailing incentive structures, training pathways, and publication norms disproportionately reward algorithmic novelty, particularly for early-career researchers, while at times undervaluing conceptual contributions that are essential for scientific maturation and clinical translation. Through representative examples from medical imaging AI, we show how insufficient conceptual grounding can lead to misaligned objectives, fragile generalization, and limited real-world impact. We conclude with actionable recommendations for researchers, mentors, reviewers, and journals to better recognize, support, and integrate conceptual innovation alongside algorithmic advances.

Tissue motion correction through image registration is essential for ultrasound localization microscopy (ULM). Parametric image registration is commonly formulated as an optimization problem where motion parameters are iteratively updated to maximize image similarity, and used optimization algorithms typically rely on gradient information, the explicit evaluation of which can become computationally demanding. This work investigates Extremum Seeking Control (ESC) as an alternative to explicit derivative evaluation in image registration. By obtaining descent information via integrating perturbed and demodulated image similarity metric across iterations, ESC avoids differentiation of the image similarity metric with respect to motion parameters in each iteration. The classical ESC, whose optimization behavior approximates that of classical gradient descent (GD), is first compared with GD for affine image registration using simulated ground-truth motions derived from a beating ex vivo porcine heart dataset. The results show that ESC achieves registration accuracy and convergence behavior comparable to GD while reducing per-iteration computational cost by approximately 3.5-fold. ESC is subsequently employed in a two-stage motion correction pipeline, where affine registration compensates for global tissue motion and B-spline registration corrects residual local deformation. The proposed method is applied to ULM imaging of a beating ex vivo porcine heart and achieves a spatial resolution of 219 um, substantially below the half-wavelength diffraction limit of 321 um associated with 2.4 MHz diverging-wave imaging. These results demonstrate that ESC provides an effective alternative to explicit derivative evaluation in ULM image registration, enabling accurate motion correction and high-quality super-resolution imaging.

CT-derived airway models support pulmonary morphometry and airflow simulation, but are often limited by distal scan resolution and the need for substantial cleanup near bifurcations. Procedural alternatives are reproducible, yet many rely on stitched tubular primitives that introduce non-smooth junctions and poorly defined open boundaries. We present RespGeomLib, a reproducible parametric engine for generating analysis-ready human airway lumen surfaces from compact YAML specifications. The framework combines port-based assembly with implicit smooth-min junction blending to produce seamless junctions, while avoiding full-tree voxelization through analytic segments and local implicit extraction around bifurcations. Quantitatively, RespGeomLib yields cleaner junctions than a Boolean/stitch baseline and is substantially faster and more memory-efficient than whole-tree global implicit extraction. We further demonstrate morphometry-guided tree generation, controlled synthetic airway variants, and CFD-ready export with stable airflow simulation. RespGeomLib targets biomedical workflows requiring reproducible morphometry, controlled synthetic variants, and simulation-ready lumen geometry. The code is publicly available at https://nichula01.github.io/Respgeomlib/

In safety-critical classification, the cost of failure is often asymmetric, yet Bayesian deep learning summarises epistemic uncertainty with a single scalar, mutual information (MI), that cannot distinguish whether a model's ignorance involves a benign or safety-critical class. We decompose MI into a per-class vector $C_k(x)=σ_k^{2}/(2μ_k)$, with $μ_k{=}\mathbb{E}[p_k]$ and $σ_k^2{=}\mathrm{Var}[p_k]$ across posterior samples. The decomposition follows from a second-order Taylor expansion of the entropy; the $1/μ_k$ weighting corrects boundary suppression and makes $C_k$ comparable across rare and common classes. By construction $\sum_k C_k \approx \mathrm{MI}$, and a companion skewness diagnostic flags inputs where the approximation degrades. After characterising the axiomatic properties of $C_k$, we validate it on three tasks: (i) selective prediction for diabetic retinopathy, where critical-class $C_k$ reduces selective risk by 34.7\% over MI and 56.2\% over variance baselines; (ii) out-of-distribution detection on clinical and image benchmarks, where $\sum_k C_k$ achieves the highest AUROC and the per-class view exposes asymmetric shifts invisible to MI; and (iii) a controlled label-noise study in which $\sum_k C_k$ shows less sensitivity to injected aleatoric noise than MI under end-to-end Bayesian training, while both metrics degrade under transfer learning. Across all tasks, the quality of the posterior approximation shapes uncertainty at least as strongly as the choice of metric, suggesting that how uncertainty is propagated through the network matters as much as how it is measured.

Although traditional statistical techniques and machine learning methods have contributed significantly to genetics and, in particular, inherited disease diagnosis, they often struggle with complex, high-dimensional data, a challenge now addressed by state-of-the-art deep learning models. Large language models (LLMs), based on transformer architectures, have excelled in tasks requiring contextual comprehension of unstructured medical data. This systematic review examines the role of LLMs in the genetic research and diagnostics of both rare and common diseases. Automated keyword-based search in PubMed, bioRxiv, medRxiv, and arXiv was conducted, targeting studies on LLM applications in diagnostics and education within genetics and removing irrelevant or outdated models. A total of 172 studies were analyzed, highlighting applications in genomic variant identification, annotation, and interpretation, as well as medical imaging advancements through vision transformers. Key findings indicate that while transformer-based models significantly advance disease and risk stratification, variant interpretation, medical imaging analysis, and report generation, major challenges persist in integrating multimodal data (genomic sequences, imaging, and clinical records) into unified and clinically robust pipelines, facing limitations in generalizability and practical implementation in clinical settings. This review provides a comprehensive classification and assessment of the current capabilities and limitations of LLMs in transforming hereditary disease diagnostics and supporting genetic education, serving as a guide to navigate this rapidly evolving field.

We present Clin-JEPA, a multi-phase co-training framework for joint-embedding predictive (JEPA) pretraining on EHR patient trajectories. JEPA architectures have enabled latent-space planning in robotics and high-quality representation learning in vision, but extending the paradigm to EHR data -- to obtain a single backbone that simultaneously forecasts patient trajectories and serves diverse downstream risk-prediction tasks without per-task fine-tuning -- remains an open challenge. Existing JEPA frameworks either discard the predictor after pretraining (I-JEPA, V-JEPA) or train it on a frozen pretrained encoder (V-JEPA 2-AC), leaving the encoder unaware of the rollout signal that the retained predictor must use at inference; co-training the encoder and predictor under a shared JEPA prediction objective would supply this grounding, but naïve co-training is unstable, with representation collapse and online/target drift causing autoregressive rollout to diverge. Clin-JEPA's five-phase pretraining curriculum -- predictor warmup, joint refinement, EMA target alignment, hard sync, and predictor finalization -- addresses each failure mode by phase, stably co-training a Qwen3-8B-based encoder and a 92M-parameter latent trajectory predictor. On MIMIC-IV ICU data, three independent evaluations support the framework: (1) latent $\ell_1$ rollout drift uniquely converges ($-$15.7%) over 48-hour horizons while baselines and ablations diverge (+3% to +4951%); (2) the encoder learns a clinically discriminative latent geometry (deteriorating-patient cohorts displace 4.83$\times$ further than stable patients in latent space, vs $\leq$2.62$\times$ for baseline encoders); (3) a single backbone outperforms strong tabular and sequence baselines on multi-task downstream evaluation. Clin-JEPA achieves mean AUROC 0.851 on ICareFM EEP and 0.883 on 8 binary risk tasks (+0.038 and +0.041 vs baseline average).

The development of medical AI is constrained by limited access to high-quality clinical data due to institutional silos and strict privacy regulations such as HIPAA and GDPR. Synthetic data generation offers a potential solution, but existing methods lack principled mechanisms to explicitly manage the privacy-utility trade-off, often degrading clinically meaningful patterns or risking patient re-identification. We present PSyGenTAB, a privacy-preserving generative framework that formulates synthetic healthcare data generation as a constrained optimization problem solved using the Augmented Lagrangian Method. By embedding configurable privacy constraints directly into model training, PSyGenTAB enforces minimum privacy thresholds while maximizing clinical data utility. Across multiple clinically motivated benchmarks, PSyGenTAB preserves inter-feature clinical relationships and minority-class diagnostic patterns essential for reliable health AI. Downstream evaluation using Train-on-Synthetic, Test-on-Real and Train-on-Real, Test-on-Synthetic protocols shows that models trained on synthetic data achieve performance comparable to those trained on real patient records. Privacy auditing further demonstrates reduced exact record reproduction and strong resilience to membership inference attacks. These results establish PSyGenTAB as a principled framework for balancing privacy protection and clinical utility in synthetic healthcare data, supporting secure cross-institutional AI development.

This paper presents the development of machine learning (ML) models to predict hypoxemia severity during emergency triage, especially in Chemical, Biological, Radiological, Nuclear, and Explosive (CBRNE) events, using physiological data from medical-grade sensors. Gradient Boosting Models (XGBoost, LightGBM, CatBoost) and sequential models (LSTM, GRU) were trained on physiological and demographic data from the MIMIC-III and IV datasets. A robust preprocessing pipeline addressed missing data, class imbalances, and incorporated synthetic data flagged with masks. Gradient Boosting Models (GBMs) outperformed sequential models in terms of training speed, interpretability, and reliability, making them well-suited for real-time decision-making. While their performance was comparable to that of sequential models, the GBMs used score features from six physiological variables derived from the enhanced National Early Warning Score (NEWS) 2, which we termed NEWS2+. This approach significantly improved prediction accuracy. While sequential models handled temporal data well, their performance gains did not justify the higher computational cost. A 5-minute prediction window was chosen for timely intervention, with minute-level interpolations standardizing the data. Feature importance analysis highlighted the significant role of mask and score features in enhancing both transparency and performance. Temporal dependencies proved to be less critical, as Gradient Boosting Models were able to capture key patterns effectively without relying on them. This study highlights ML's potential to improve triage and reduce alarm fatigue. Future work will integrate data from multiple hospitals to enhance model generalizability across clinical settings.

Accurate survival prediction is essential for personalized treatment planning in head and neck cancer, yet remains challenging due to the heterogeneous and high-dimensional nature of multimodal clinical data. While deep survival models have improved predictive performance over classical statistical approaches, existing methods typically rely on static fusion strategies or temporally agnostic modeling, limiting their ability to capture structured clinical workflows. In this work, we propose ChronoSurv, a heterogeneous hierarchical directed graph framework for multimodal survival analysis. ChronoSurv represents patient care as a progression-aware clinical trajectory using directed graphs aligned with key diagnostic steps. A hierarchical topology incorporates fine-grained, coarse, and global representations, further supporting flexible adaptation to missing modalities, while heterogeneous message passing models complex and asymmetric relationships across modalities and clinical steps. Experimental results on two public datasets demonstrate that ChronoSurv achieves state-of-the-art discriminative performance while maintaining statistically reliable calibration. Comprehensive ablation studies further confirm the contribution of each architectural component, highlighting the potential of trajectory-aware graph modeling for multimodal survival prediction.

Mild Cognitive Impairment (MCI) is a medical condition characterized by a noticeable decline in memory, language, or thinking abilities. MCI detection from spontaneous speech is promising for scalable screening. However, learned models often exploit demographic cues correlated with labels, resulting in a large performance gap across subgroups. We present a multimodal framework that combines (i) cross-model fusion between modalities (speech, text, and image), and (ii) unlearning using gradient reversal that discourages the shared embedding from encoding task-irrelevant demographic attributes. Evaluated on the multilingual benchmarks TAUKADIAL and PREPARE, our method outperforms the state-of-the-art multilingual and multimodal baseline in MCI classification while substantially reducing the performance gap across patient subgroups (sex and language). We further analyze transfer across datasets, showing that demographic unlearning helps learn more robust representations for MCI detection.

Asthma causes expiratory airflow limitation and is clinically assessed using spirometry, which provides the FEV1/FVC ratio representing the proportion of air exhaled in the first second relative to total forced vital capacity. Prior studies suggest that respiratory sounds recorded at posterior sites (Left Lower, Left Upper, Right Upper, Right Lower) reflect regional airflow patterns. In this study, we investigate the relationship between the expiratory-to-inspiratory (E/I) spectral power ratio and FEV1/FVC in 141 participants aged 20-60 years using Spearman correlation across frequency subbands. The 100-200 Hz and 200-400 Hz bands showed significant correlations. Overall, lower posterior sites showed stronger associations; younger adults showed stronger correlations at the Left Lower site, whereas older adults showed stronger correlations at the Left Upper site. Gender-stratified analysis showed stronger Left Lower correlations in males and stronger Left Upper correlations in females.

Accurate disease risk prediction is challenged by heterogeneous features, limited data, and class imbalance. This study presents yvsoucom-iterkit, a deterministic AutoML framework that models pipeline optimization as a configuration-level system with full reproducibility and traceable execution logs, enabling systematic analysis of component attribution, interactions, similarity, and cross-seed robustness. Experiments on the Pima Indians Diabetes and Stroke datasets across more than 18,000 pipeline configurations reveal a structured yet partially redundant search space, where performance is dominated by a small subset of interacting components. Ensemble models achieve stable performance, reaching a Weighted-F1 of 0.89 on Pima and 0.94 on Stroke. Macro-F1 reaches approximately 0.88 on Pima but drops to 0.6560 on Stroke due to severe imbalance. Cross-seed experiments show that ensembles reduce variance compared to single models. Friedman testing ($p < 0.05$) confirms significant ranking differences across configurations. Based on analysis of component attribution, interaction, and similarity, optimal configuration design reveals dataset-dependent behavior. For the Pima dataset, computational efficiency benefits from simplified search spaces where redundant components can be removed, with split ratio playing a key role. In contrast, the Stroke dataset requires enhanced imbalance-aware strategies, where RandomOverSampler improves Macro-F1 from 0.6560 to 0.6766. These findings demonstrate that effective AutoML optimization is achieved through optimal configuration design, where carefully constraining the search space to high-impact components can improve performance, stability, and interpretability while reducing unnecessary search complexity.

Dysarthric speech quality assessment (DSQA) is critical for clinical diagnostics and inclusive speech technologies. However, subjective evaluation is costly and difficult to scale, and the scarcity of labeled data limits robust objective modeling. To address this, we propose a three-stage framework that leverages unlabeled dysarthric speech and large-scale typical speech datasets to scale training. A teacher model first generates pseudo-labels for unlabeled samples, followed by weakly supervised pretraining using a label-aware contrastive learning strategy that exposes the model to diverse speakers and acoustic conditions. The pretrained model is then fine-tuned for the downstream DSQA task. Experiments on five unseen datasets spanning multiple etiologies and languages demonstrate the robustness of our approach. Our Whisper-based baseline significantly outperforms SOTA DSQA predictors such as SpICE, and the full framework achieves an average SRCC of 0.761 across unseen test datasets.

Externally controlled trials compare outcomes from a single-arm trial with external controls drawn from historical trials, registries, or observational studies. For time-to-event endpoints, a key challenge arises when follow-up is indexed at treatment initiation in the single-arm trial, but the external-control data are indexed at an earlier clinical milestone, such as diagnosis or relapse. This misalignment can induce immortal time bias, distort risk sets, and complicate the interpretation of survival comparisons. We propose Index Date Imputation (IDI), a truncation-aware approach for imputing comparable index dates for external-control patients in settings with delayed treatment initiation. IDI estimates the marginal distribution of treatment-initiation times in the target single-arm population while accounting for the fact that initiation times are observed only among patients who survive long enough to initiate treatment. The imputed index dates are then used to align follow-up and enforce comparable truncation conditions in the external-control cohort. Because temporal alignment alone does not address population-level confounding, IDI is combined with propensity score weighting or matching to improve covariate comparability between cohorts. We evaluate the finite-sample performance of the proposed approach through Monte Carlo simulation studies. Using data from a randomized oncology trial, we emulate an externally controlled analysis with induced index-date misalignment and show that IDI reduces discrepancy from the randomized trial benchmark. IDI provides a practical strategy for index-date alignment in survival analyses involving delayed treatment initiation and can be integrated with standard covariate-adjustment methods when suitable external controls are available.

Glucose forecasting algorithms are an important aspect of glycemic control management in type 1 diabetes. So far, the research community has developed numerous algorithms and models for forecasting. However, it is well-recognized that the lack of standardized model performance evaluation benchmarks makes fair comparison difficult and hinders further innovation, and thus benchmark standardization is in urgent need. Furthermore, many published glucose forecasting algorithms are limited to CGM data alone, ignoring other multimodal signals such as insulin dosing and carbohydrate intake. Here, we introduce MetaboNet-Bench, a benchmark for multimodal glucose forecasting for patients with type 1 diabetes that provides an extensible open-source evaluation framework for comparison of glucose forecasting algorithms that leverage glucose, insulin, and carbohydrate data. We then demonstrate its utility by benchmarking several recently published glucose forecasting models and a custom multimodal time-series model, representing different model architectures. The results show that the benefit of adding data modalities is conditioned on the complexity of the model and that incorporating more clinical metrics helps identify meaningful gaps to fill for future research.

Forces transmitted by bones are routinely studied in human biomechanics, but it is challenging to measure them non-invasively, especially outside of laboratory settings. We introduce a technique for non-invasive, in vivo measurement of tibial compressive force using flexural waves propagating in the tibia. Modelling the tibia as an axially compressed Euler-Bernoulli beam, we show that tibial flexural waves have load-dependent frequency spectra. Specifically, under physiological conditions, peak locations in the wave acceleration spectra vary linearly with the compressive force on the tibia and may be used as proxies for the compressive force. We test the validity of this technique using a proof-of-concept wearable system that generates flexural waves via a skin-mounted mechanical transducer and measures the spectra of these waves using a skin-mounted accelerometer. In agreement with beam theory, data from 9 participants demonstrate linear relationships between tibial compressive force and spectral peak location, with Pearson correlation coefficients $r=0.82 - 0.99$ (mean $r=0.93$) for medial-lateral swaying and $r=0.81 - 0.98$ (mean $r=0.93$) for walking trials. This flexural wave-based technique could give rise to a new class of wearable sensors for non-invasive physiological bone load monitoring and measurement, impacting research in human locomotion and sports medicine.

Psychophysical discrimination of structured light (SL) stimuli may be useful in screening for various macular disorders, including degenerative macular diseases. The circularly-oriented macular pigment optical density (coMPOD), calculated from the discrimination performance of SL-induced entoptic phenomena, may reveal a novel functional biomarker of macular health. In this study, we investigated the potential influence of eye dominance and testing order effects on SL-based stimulus perception, factors that potentially influence the sensitivity of screening tests based on SL technology. A total of 28 participants (aged 18-38 years) were selected for the study after undergoing a comprehensive eye examination. A psychophysical task was performed where various SL-based entoptic images with multiple azimuthal fringes rotating with a specific temporal frequency were projected onto the participants' retinas. By occluding the central areas of entoptic images, we measured the retinal eccentricity ($R$) of the perceivable area of the stimuli. The slope of the coMPOD profile ($a$-value) was calculated for each participant using a spatiotemporal sensitivity model that takes into account the perceptual threshold measurements of structured light stimuli with varying spatial densities and temporal frequencies. The Pearson correlation coefficient between eye dominance and testing order effects was $r=0.8$ ($p<0.01$). The Bland-Altman plots for both factors indicated zero bias. The results indicate repeatable measurements for both eyes, implying minimal impact from eye dominance and testing order on SL-based stimulus perception. The results provide a foundation for future studies exploring the clinical utility of SL tools in eye health.