SV-Cine: Diagnosis-Conditioned Segmentation of Single Ventricle Physiology via Generative Data Augmentation

Abstract

Single Ventricle Physiology (SVP) is a rare subtype of congenital heart disease characterized by the presence of a single functional cardiac ventricle with atypical anatomic configurations that challenge conventional image segmentation approaches. The scarcity of clinical data and the morphological diversity across SVP subtypes make the development of robust segmentation methods particularly difficult. To address these limitations, we propose a cardiac MRI segmentation framework focused on ventricular chambers and myocardium segmentation tailored for SVP. First, we introduce a data augmentation pipeline that generates synthetic 3D cardiac meshes using SDF4CHD and corresponding synthetic cardiac MRI through generative modeling. Second, we introduce SV-Cine, a diagnosis-conditioned adaptation of the foundation model CineMA that incorporates patient-level diagnostic information through Feature-wise Linear Modulation layers, enabling diagnosis-aware feature adaptation during segmentation. We evaluated the framework on an internal cohort with varying SVP subtypes. SV-Cine achieved median Dice scores of 0.89 (IQR: 0.80--0.91) for the left ventricle and 0.72 (IQR: 0.54--0.84) for the right ventricle, outperforming the strongest baseline, nnU-Net, by 0.39 Dice points on right ventricle segmentation. It also yields a median ejection fraction error of 5.55 percentage points (IQR: 3.41--7.69) for the dominant ventricle. Compared with the internal cohort, LV and myocardium segmentation performance was lower for the external cohort; whereas RV Dice scores were comparable for both cohorts. Our findings suggest that a pretrained foundation model can be adapted for highly specialized downstream tasks through usage of diagnosis priors while leveraging anatomic knowledge learned from large-scale MRI datasets during pretraining.

Hybrid Machine Learning-Assisted Raman Spectroscopy with Generative Feature Augmentation for Pharmaceutical Identification
Quach Thi Thai Binh, Ton Nu Quynh Trang, Thang B. Phan +2
arXiv1 Oct 2026PreprintCC BY 4.0Full textCondensed matterarXiv
Recovering molecules from coarse-grained beads: free-energy-conditioned generative backmapping across chemical space
Luis Itza Vazquez-Salazar, Tristan Bereau
arXiv3 Sep 2026PreprintCC BY-NC-SA 4.0Full textPhysical chemistry Biological & medical physicsarXiv
A target-domain augmentation and class-wise alignment framework for few-shot cross-condition bearing fault diagnosis
Yijia Hu, Ying Gu, Yanjie Su +1
PLOS One29 Oct 2026Peer-reviewedCC BYFull textBioinformatics & computational biology Evolutionary biologyPLOSEurope PMCPMC
Evaluation of a markerless data augmentation algorithm for sports movements
Axel Koussou, Hanane Ariouat, Sebastien Laporte +1
Multidisciplinary Biomechanics Journal26 Oct 2026Peer-reviewedCC BY 4.0Full textGeneral physicsHAL
Augmentation curve for torus knots
Alexander Hock, Sergey Shadrin
arXiv15 Oct 2026PreprintarXiv licenceFull textAlgebra & number theory Geometry & topologyarXiv

Cite

BibTeX RIS

BibTeX

@misc{cunge2026svcine,
  title = {{SV-Cine: Diagnosis-Conditioned Segmentation of Single Ventricle Physiology via Generative Data Augmentation}},
  author = {Lila Cunge},
  howpublished = {arXiv},
  year = {2026},
  month = sep,
  doi = {10.48550/arxiv.2609.12997},
  eprint = {2609.12997},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.12997},
}

RIS

TY  - UNPB
TI  - SV-Cine: Diagnosis-Conditioned Segmentation of Single Ventricle Physiology via Generative Data Augmentation
AU  - Lila Cunge
PY  - 2026
DA  - 2026/09/11
PB  - arXiv
DO  - 10.48550/arxiv.2609.12997
UR  - https://arxiv.org/abs/2609.12997
AB  - Single Ventricle Physiology (SVP) is a rare subtype of congenital heart disease characterized by the presence of a single functional cardiac ventricle with atypical anatomic configurations that challenge conventional image segmentation approaches. The scarcity of clinical data and the morphological diversity across SVP subtypes make the development of robust segmentation methods particularly difficult. To address these limitations, we propose a cardiac MRI segmentation framework focused on ventricular chambers and myocardium segmentation tailored for SVP. First, we introduce a data augmentation pipeline that generates synthetic 3D cardiac meshes using SDF4CHD and corresponding synthetic cardiac MRI through generative modeling. Second, we introduce SV-Cine, a diagnosis-conditioned adaptation of the foundation model CineMA that incorporates patient-level diagnostic information through Feature-wise Linear Modulation layers, enabling diagnosis-aware feature adaptation during segmentation. We evaluated the framework on an internal cohort with varying SVP subtypes. SV-Cine achieved median Dice scores of 0.89 (IQR: 0.80--0.91) for the left ventricle and 0.72 (IQR: 0.54--0.84) for the right ventricle, outperforming the strongest baseline, nnU-Net, by 0.39 Dice points on right ventricle segmentation. It also yields a median ejection fraction error of 5.55 percentage points (IQR: 3.41--7.69) for the dominant ventricle. Compared with the internal cohort, LV and myocardium segmentation performance was lower for the external cohort; whereas RV Dice scores were comparable for both cohorts. Our findings suggest that a pretrained foundation model can be adapted for highly specialized downstream tasks through usage of diagnosis priors while leveraging anatomic knowledge learned from large-scale MRI datasets during pretraining.
ER  -

Abstract

Related

Cite