Cross-Anatomy Transfer Versus Sparse Interpolation in Digital-Twin-Oriented Aortic Fluid-Structure Interaction Surrogates

Nourbakhsh, Ali; Niroomand, Mohammad Reza; Nourbakhsh, Erfan

Under Review · ICBME 2026

Cross-Anatomy Transfer Versus Sparse Interpolation in Digital-Twin-Oriented Aortic Fluid-Structure Interaction Surrogates

Ali Nourbakhsh¹, Mohammad Reza Niroomand¹, Erfan Nourbakhsh²

¹Isfahan University of Technology
²University of Isfahan

33rd National and 11th International Iranian Conference on Biomedical Engineering (ICBME 2026)

arXiv Code BibTeX

Paper

Abstract

Surrogate credibility for fluid-structure interaction (FSI) requires distinguishing transfer across independent anatomies from interpolation within an already sampled surface. Four de-identified human aortic models from the Vascular Model Repository were reconstructed into separate lumen and nominal 1.5-mm wall domains and analyzed under matched first-cycle two-way FSI. A geometry-only LightGBM prior, selected by leave-one-anatomy-out development on three anatomies, was zero-shot evaluated on a fourth, then probed with a post-zero-shot sparse field-completion case study over six targets. Zero-shot transfer was poor across all targets. At a five-percent anchor level (203 anchors, 3,852 evaluation nodes), prior-plus-adaptation reached an oscillatory shear index (OSI) R² of 0.603. However, same-anchor controls tuned only on the three development anatomies were stronger for several outcomes: inverse-distance weighting reached R² = 0.829 (OSI), 0.617 (peak von Mises stress), 0.676 (mean stress); radial basis function interpolation reached 0.917, 0.714, 0.778. Sparse within-anatomy labels thus support field completion, but this four-anatomy cohort gives no evidence the cross-anatomy prior adds value beyond direct interpolation. We frame this as a first computational stage toward a measurement-linked digital twin: the surrogate/update layer is evaluated here, while larger cohorts, converged FSI, measurable patient-side inputs, and physics-informed learning remain future work, not a claim of a complete clinical twin.

Materials and Methods

Reconstructed Aortic Geometries

Four VMR aortic models were reconstructed into separate lumen and wall domains under a deterministic CAD workflow, then solved with matched two-way FSI assumptions.

Figure 1. Reconstructed geometries for development anatomies A1-A3, matched display scale after principal-axis alignment. Opaque: blood-lumen/fluid domain; translucent: CATIA-generated wall domain. Renderings, not patient photographs.

Metric	A1	A2	A3	A4
Wall nodes	8191	6800	3203	4055
Tetrahedra	278889	213130	415816	128656
Min tet. quality	.145	.077	.121	.144
Pseudo-centerline length (mm)	128.09	96.21	93.69	85.40
Tortuosity (geom.)	1.226	1.256	1.138	1.346
Mean TAWSS (Pa)	.964	1.218	.699	1.144
P95 TAWSS (Pa)	2.380	4.559	1.094	2.878
Global saved-state peak WSS (Pa)	74.73	134.27	33.31	101.79
P95 saved-state peak WSS (Pa)	14.14	27.41	5.04	15.55
Mean saved-state max VM (kPa)	3.916	2.472	3.140	2.518
P95 saved-state max VM (kPa)	10.55	4.76	5.92	4.87

Table 1. Computational anatomy and FSI summary. Pseudo-centerline length/tortuosity are geometry-descriptor values, not clinical measurements.

Study Design

Evaluation Workflow

Development uses leave-one-anatomy-out on A1-A3. Anatomy 4 is held out for zero-shot transfer and a post-zero-shot sparse-calibration case study.

Figure 2. Evaluation workflow and research roadmap. Solid boxes: capabilities evaluated here. Dashed boxes: planned extensions, not current results.

Results

Calibration and Prior-Plus-Adaptation

Zero-shot transfer is weak. With 5% anchors on anatomy 4, prior-plus-adaptation recovers useful structure for some targets (OSI R² = 0.603), but same-anchor baselines remain critical controls.

LOAO calibration behavior across six primary targets.

Figure 3. Development-only LOAO calibration behavior for the six primary targets (R², Spearman ρ, top-decile overlap). Vertical line: preselected 5% setting.

Target	R²	MAE	RMSE	ρ	Top10	Emp. band
TAWSS (Pa)	.234	.408	1.179	.671	.587	.884
OSI	.603	.0438	.0645	.807	.668	.914
RRT (Pa⁻¹)	.000	5.959	287.697	.711	.579	.900
Saved-state peak WSS (Pa)	.302	2.035	5.835	.732	.605	.894
Saved-state max VM (kPa)	.317	.729	1.029	.569	.506	.903
Temporal-mean VM (kPa)	.410	.170	.239	.673	.506	.901

Table 2. Prior-plus-adaptation performance on anatomy 4, 5% setting (non-anchor nodes only).

Results: Same-Anchor Controls

Does the Prior Add Value?

Direct IDW/RBF interpolation with the same anchors often matches or beats prior-plus-adaptation. Sparse labels help, but credit must not go automatically to cross-anatomy transfer.

Same-anchor controls comparing prior-plus-adaptation to IDW and RBF.

Figure 4. Same-anchor controls on anatomy 4 at 5% calibration. Direct IDW/RBF interpolation frequently equal or exceed prior-plus-adaptation.

Target	Prior+adapt		KNN anchors		IDW anchors		RBF anchors
Target	R²	ρ	R²	ρ	R²	ρ	R²	ρ
TAWSS	.234	.671	.228	.679	.185	.747	.133	.785
OSI	.603	.807	.664	.835	.829	.899	.917	.950
RRT	.000	.711	.000	.731	.000	.783	.000	.835
Saved-state peak WSS	.302	.732	.299	.747	.277	.784	.013	.815
Saved-state max VM	.317	.569	.359	.573	.617	.754	.714	.823
Temporal-mean VM	.410	.673	.439	.677	.676	.832	.778	.878

Table 3. Anatomy-4 reconstruction at 5% calibration: prior-plus-adaptation vs. direct baselines. RRT R² shown as 0.000 to avoid negative-zero formatting.

Results: Spatial Fields

Field Reconstruction Maps

Spatial maps reinforce the table findings: RBF reconstructs OSI and maximum VM well, while peak WSS remains harder and unconstrained RBF can produce a few out-of-range OSI values.

Anatomy-4 spatial field comparisons across methods.

Figure 5. Anatomy-4 non-anchor spatial fields at 5% calibration. Rows: OSI, saved-state peak WSS, saved-state maximum VM stress. Columns: FSI reference, prior-plus-adaptation, IDW, RBF; matched display limits per row.

Results: Diagnostics

Residual Bands and Geometry Shift

Development-calibrated residual bands and local geometry-shift scores are diagnostic markers, not guarantees of reconstruction error.

Figure 6. Secondary diagnostics on non-anchor anatomy-4 nodes: empirical residual-band width for OSI (normalized by development IQR) and local geometry-shift percentile. Not clinically calibrated.

BibTeX

@misc{nourbakhsh2026aorticfsisurrogates,
  title={Cross-Anatomy Transfer Versus Sparse Interpolation in Digital-Twin-Oriented Aortic Fluid-Structure Interaction Surrogates},
  author={Ali Nourbakhsh and Mohammad Reza Niroomand and Erfan Nourbakhsh},
  year={2026},
  eprint={2609.16322},
  archivePrefix={arXiv},
  primaryClass={cs.AI},
  url={https://arxiv.org/abs/2609.16322},
  note={Under review at ICBME 2026},
}