Time-resolved aortic 3D shape reconstruction from a limited number of cine 2D MRI slices
arXiv:2602.11873v2 Announce Type: replace Abstract: Background and Objective: To assess the feasibility and accuracy of reconstructing time-resolved, three-dimensional, subject-specific aortic geometries from a limited number of standard cine 2D magnetic resonance imaging (MRI) acquisitions. This is achieved by coupling a statistical shape model with a differentiable volumetric mesh optimization algorithm. Methods: Cine 2D MRI slices were manually segmented and used to reconstruct subject-specific aortic geometries via a differentiable mesh optimization algorithm, constrained by a statistical shape model. Optimal slice positioning was first evaluated on synthetic data, followed by in-vivo acquisition in 30 subjects (19 volunteers and 11 aortic stenosis patients). Time-resolved aortic geome
View PDF
Abstract:Background and Objective: To assess the feasibility and accuracy of reconstructing time-resolved, three-dimensional, subject-specific aortic geometries from a limited number of standard cine 2D magnetic resonance imaging (MRI) acquisitions. This is achieved by coupling a statistical shape model with a differentiable volumetric mesh optimization algorithm. Methods: Cine 2D MRI slices were manually segmented and used to reconstruct subject-specific aortic geometries via a differentiable mesh optimization algorithm, constrained by a statistical shape model. Optimal slice positioning was first evaluated on synthetic data, followed by in-vivo acquisition in 30 subjects (19 volunteers and 11 aortic stenosis patients). Time-resolved aortic geometries were reconstructed, from which geometric descriptors and radial strain were derived. In a subset of 10 subjects, 4D flow MRI data was acquired to provide volumetric reference for peak-systolic shape comparison. Results: Accurate reconstruction was achieved using as few as six cine 2D MRI slices. Agreement with 4D flow MRI reference data yielded a Dice score of (89.9 +/- 1.6) %, Intersection over Union of (81.7 +/- 2.7) %, Hausdorff distance of (7.3 +/- 3.3) mm, and Chamfer distance of (3.7 +/- 0.6) mm. The mean absolute radius error along the aortic arch was (0.8 +/- 0.6) mm. Secondary analysis demonstrated significant differences in geometric features and radial strain across age groups, with strain decreasing progressively with age at values of (11.00 +/- 3.11) x 10-2 vs. (3.74 +/- 1.25) x 10-2 vs. (2.89 +/- 0.87) x 10-2 for the young, mid-age, and elderly groups, respectively. Conclusion: The proposed framework enables reconstruction of time-resolved, subject-specific aortic geometries from a limited number of standard cine 2D MRI acquisitions, providing a practical basis for downstream computational analysis.
Subjects:
Image and Video Processing (eess.IV); Medical Physics (physics.med-ph); Methodology (stat.ME)
Cite as: arXiv:2602.11873 [eess.IV]
(or arXiv:2602.11873v2 [eess.IV] for this version)
https://doi.org/10.48550/arXiv.2602.11873
arXiv-issued DOI via DataCite
Submission history
From: Gloria Wolkerstorfer [view email] [v1] Thu, 12 Feb 2026 12:23:41 UTC (1,277 KB) [v2] Tue, 31 Mar 2026 14:44:35 UTC (1,737 KB)
Sign in to highlight and annotate this article
Conversation starters
Daily AI Digest
Get the top 5 AI stories delivered to your inbox every morning.
More about
modelannouncefeature
93% of a Claude Code Session Is Noise. Here's the Proof.
I wrote a session distiller for Claude Code that cuts 70MB sessions down to 7MB. That post covered the what and how. This one covers the why, with data. Before I wrote a single line of code, I needed to answer three questions: What's actually inside a 70MB session? What's safe to throw away? How do I prove I'm not losing anything that matters? Dissecting a 70MB Session I opened a real 70MB session JSONL and categorized every byte. Here's the breakdown: JSON envelope (sessionId, cwd, version, gitBranch): ~54% Tool results (Read, Bash, Edit, Write, Agent): ~25% Base64 images (screenshots, UI captures): ~12% Thinking blocks (internal reasoning): ~4% Actual conversation text: ~3% Progress lines, file-history-snapshots: ~2% That first line is the surprise. Every single JSONL line repeats the sa
Can AI Predict Market Crashes Better Than Human Experts? The Data-Driven Verdict for 2024
Can AI Predict Market Crashes Better Than Human Experts? The Data-Driven Verdict for 2024 In a world grappling with persistent inflation, shifting interest rates, and looming recession fears, the allure of AI predicting market crashes is stronger than ever. While Indian IT firms and global tech giants promise unprecedented efficiency gains through AI, investors are increasingly asking if artificial intelligence can truly offer foresight beyond human capabilities, fundamentally altering investment strategies NOW. The data-driven verdict suggests AI excels at pattern recognition and processing vast amounts of information, but its ability to predict 'black swan' events or truly understand human irrationality remains a significant challenge, making it a powerful tool for augmentation rather th
Knowledge Map
Connected Articles — Knowledge Graph
This article is connected to other articles through shared AI topics and tags.
More in Models
Same Instruction File, Same Score, Completely Different Failures
Two AI coding agents were given the same task with the same 10-rule instruction file. Both scored 70% adherence. Here's the breakdown: Rule Agent A Agent B camelCase variables PASS FAIL No any type FAIL PASS No console.log FAIL PASS Named exports only PASS FAIL Max 300 lines PASS FAIL Test files exist FAIL PASS Agent A had a type safety gap. It used any for request parameters even though it defined the correct types in its own types.ts file. Agent B had a structural discipline gap. It used snake_case for a variable, added a default export following Express conventions over the project rules, and generated a 338-line file by adding features beyond the task scope. Same score. Completely different engineering weaknesses. That table came from RuleProbe . About this case study The comparison us
SQUIRE: Interactive UI Authoring via Slot QUery Intermediate REpresentations
Frontend developers create UI prototypes to evaluate alternatives, which is a time-consuming process of repeated iteration and refinement. Generative AI code assistants enable rapid prototyping simply by prompting through a chat interface rather than writing code. However, while this interaction gives developers flexibility since they can write any prompt they wish, it makes it challenging to control what is generated. First, natural language on its own can be ambiguous, making it difficult for developers to precisely communicate their intentions. Second, the model may respond unpredictably…
An Implementation Guide to Running NVIDIA Transformer Engine with Mixed Precision, FP8 Checks, Benchmarking, and Fallback Execution
In this tutorial, we implement an advanced, practical implementation of the NVIDIA Transformer Engine in Python, focusing on how mixed-precision acceleration can be explored in a realistic deep learning workflow. We set up the environment, verify GPU and CUDA readiness, attempt to install the required Transformer Engine components, and handle compatibility issues gracefully so that [ ] The post An Implementation Guide to Running NVIDIA Transformer Engine with Mixed Precision, FP8 Checks, Benchmarking, and Fallback Execution appeared first on MarkTechPost .
Discussion
Sign in to join the discussion
No comments yet — be the first to share your thoughts!