The paper introduces a latent dynamical model that captures full‑cycle biventricular cardiac motion from cine MRI using a heart‑rate‑aware neural ODE and a graph‑based mesh autoencoder. It encodes ventricular anatomy and motion as a continuous latent trajectory, with a covariate‑conditioned prior for the end‑diastolic state and a Cox model linking deviations to incident heart failure. In a large UK Biobank cohort, adding the latent score to existing risk equations improved predictive performance, outperforming several established cardiac markers and alternative modeling approaches.
arXiv:2606. 26718v1 Announce Type: new Abstract: Cardiac magnetic resonance imaging (CMR) captures rich spatiotemporal information about ventricular structure and motion, but conventional risk models use only a few image-derived indices from selected cardiac phases.
arXiv:2610.09185v1 Announce Type: new Abstract: Cine cardiovascular magnetic resonance (CMR) captures the cardiac cycle as a four-dimensional (4D) sequence, but standard acquisition requires electroc...
arXiv:2610.09397v1 Announce Type: new Abstract: Cine cardiovascular magnetic resonance (CMR) analysis relies on multi-frame sequences capturing the full cardiac cycle. However, standard multi-frame a...
arXiv:2608. 09460v1 Announce Type: new Abstract: Cardiac digital twin research is moving from subject-specific anatomical replicas toward virtual cohorts that represent clinically relevant population subgroups.
CardioDrift is a fully automatic, atlas‑based framework that tracks cardiac motion on a 3D atlas mesh using a shape‑aware Coherent Point Drift algorithm. It jointly registers atlas surfaces of multiple cardiac regions to cine CMR segmentation contours across the cardiac cycle, while embedding slice‑shift correction to reduce misalignment artefacts. In evaluations on 1,301 UK Biobank subjects and the M&M dataset, CardioDrift achieved Dice similarity coefficients of 92.23 ± 8.70 % for the left ventricle, 82.11 ± 7.56 % for the myocardium, and 88.69 ± 4.13 % for the right ventricle, outperforming all baselines on every distance and overlap metric.
arXiv:2608. 19738v1 Announce Type: cross Abstract: Full-cycle biventricular geometry is essential for characterizing cardiac function.
Many clinical prediction models treat post-intervention outcomes as a one-step mapping from baseline measurements to a future endpoint. However, recovery after a procedure often unfolds as an irregular trajectory: clinical observations, medication changes, repeat interventions, and physiological measurements are recorded asynchronously and can change risk assessment over time.
arXiv:2608. 13518v1 Announce Type: new Abstract: Many clinical prediction models treat post-intervention outcomes as a one-step mapping from baseline measurements to a future endpoint.
The paper introduces SV-Cine, a cardiac MRI segmentation framework tailored for single ventricle physiology (SVP). It combines a generative data augmentation pipeline that creates synthetic 3D cardiac meshes and MRI, with a diagnosis-conditioned adaptation of the CineMA foundation model that uses patient-level diagnostic information to improve segmentation. Evaluations on an internal cohort show high Dice scores for left and right ventricles, outperforming nnU-Net, and demonstrate that incorporating diagnosis priors can adapt a pretrained model to specialized SVP tasks.
arXiv:2606. 00123v1 Announce Type: cross Abstract: Multimodal Large Language Models (MLLMs) have shown strong performance on public medical benchmarks, yet existing evaluations often remain weak proxies for clinical use, relying on isolated inputs and simplified recognition-style tasks.
arXiv:2606. 26764v1 Announce Type: cross Abstract: Developing robust artificial intelligence models for 4D (3D + time) medical imaging is constrained by limited annotated data, inter-device domain shifts, and privacy restrictions.
The paper introduces ORBIT, a self‑supervised method for detecting end‑diastolic and end‑systolic cardiac phases in fetal echocardiography without manual annotations. ORBIT learns a latent motion trajectory through registration, enabling orientation‑robust identification of phase transitions across diverse fetal heart positions. Evaluated on normal and congenital heart disease cases, it achieves low mean absolute errors (≈1.9–2.4 frames) and outperforms prior annotation‑free approaches that assume fixed orientations.