Organizations: School of Mathematical Sciences, Ocean University of China, Qingdao, 266100, Shandong, China. · Department of Cardiology, Qilu Hospital (Qingdao), Cheeloo College of Medicine, Shandong University, No. 758 Hefei Road, Qingdao, 266000, Shandong, China.
Conventional pixel-wise loss functions fail to enforce topological constraints in coronary vessel segmentation, producing fragmented vascular trees despite high pixel-level accuracy. We present ARIADNE, a two-stage framework coupling preference-aligned perception with RL-based diagnostic reasoning for topologically coherent stenosis detection. The perception module employs DPO to fine-tune the Sa2VA vision-language foundation model using Betti number constraints as preference signals, aligning the policy toward geometrically complete vessel structures rather than pixel-wise overlap metrics. The reasoning module formulates stenosis localization as a Markov Decision Process with an explicit rejection mechanism that autonomously defers ambiguous anatomical candidates such as bifurcations and vessel crossings, shifting from coverage maximization to reliability optimization. On 1,400 clinical angiograms, ARIADNE achieves state-of-the-art centerline Dice of 0.838, reduces false positives by 41% compared to geometric baselines. External validation on multi-center benchmarks ARCADE and XCAD confirms generalization across acquisition protocols. This represents the first application of DPO for topological alignment in medical imaging, demonstrating that preference-based learning over structural constraints mitigates topological violations while maintaining diagnostic sensitivity in interventional cardiology workflows.
Figures & tables
Figure 1: Training framework of Anatomy-Aware Segmentation
Figure 2: Training framework of Structure-Guided Reasoning
Figure 3: Performance comparison of our model at different stages
Method
IoU
Acc
Pre
Sen
clDice
NSD
Dice
MedSAM3 Liu et al. (2025)
0.5612
0.9650
0.7015
0.7320
0.7105
0.5821
0.7189
UNet Ronneberger et al. (2015)
0.6321
0.9798
0.7823
0.7712
0.7987
0.6478
0.7734
UNet++ Zhou et al. (2018)
0.6456
0.9805
0.7912
0.7798
0.8056
0.6545
0.7845
FR-Unet Liu et al. (2022)
0.6534
0.9815
0.7998
0.7865
0.8145
0.6656
0.7905
H-vmunet Wu et al. (2025)
0.6589
0.9820
0.8034
0.7912
0.8212
0.6712
0.7945
SVSNet Bai et al. (2025)
0.6612
0.9822
0.8067
0.7945
0.8245
0.6756
0.7960
Table 1: Comparative performance of segmentation methods on the in-house dataset (n=140). Bold indicates best performance.
Method
IoU
Acc
Pre
Sen
clDice
NSD
Dice
MedSAM3 Liu et al. (2025)
0.4532
0.9315
0.5521
0.6845
0.6215
0.3842
0.6237
UNet Ronneberger et al. (2015)
0.5234
0.9532
0.6234
0.8134
0.7321
0.4567
0.6987
UNet++ Zhou et al. (2018)
0.5356
0.9556
0.6312
0.8189
0.7412
0.4678
0.7045
H-vmunet Wu et al. (2025)
0.5412
0.9578
0.6356
0.8212
0.7489
0.4734
0.7089
FR-Unet Tian et al. (2025)
0.5456
0.9585
0.6389
0.8245
0.7523
0.4789
0.7123
SVSNet Bai et al. (2025)
0.5489
0.9592
0.6412
0.8278
0.7567
0.4812
0.7156
Table 2: External validation performance on XCAD dataset (n=126) Du et al. (2021) . Bold indicates best performance.
Figure 4: Qualitative spatiotemporal consistency analysis across the full angiographic sequence. Columns represent different models, while rows illustrate the hemodynamic progression from Wash-in (top) to Peak (middle) and Wash-out (bottom) phases. The foundation model MedSAM3 Liu et al. (2025) (Column c) exhibits significant topological fragmentation during the low-contrast wash-out phase (red arrows), confirming the semantic-topological gap. In contrast, ARIADNE (Column j) maintains robust structural continuity throughout the sequence (green arrows).
Method
TPR (Recall)
PPV (Precision)
F1 Score
FPPI ↓
Stenunet Lin et al. (2023)
0.812
0.557
0.660
2.45
LT-YOLO Li et al. (2025)
0.729
0.628
0.692
1.89
DeepDiscern Du et al. (2021)
0.773
0.588
0.667
2.12
ARIADNE
0.867
0.634
0.732
0.85
Table 3: Comparative stenosis detection performance. Bold indicates best performance per metric.
Figure 5: Each row represents a different clinical case. Left Column: Original X-ray angiograms. Middle Column: The extracted vascular tree with detected stenosis locations (marked by blue dots for candidates and green dots for final detections) identified by the RL navigation agent. Right Column: Expert annotations highlighting the ground truth stenotic lesions (indicated by red arrows).The alignment between the agent’s predictions and expert labels demonstrates the system’s capability to accurately localize hemodynamically significant lesions even in complex anatomical configurations.
Artificial Intelligence and Robotics Innovation Center, China Medical University Hospital, Taichung, Taiwan · Division of Cardiovascular Medicine, Department of Internal Medicine, China Medical University Hospital, Taichung, Taiwan · Department of Neurology, China Medical University Hospital, China Medical University, Taichung, Taiwan +2
Computer Aided Medical Procedures, Technical University of Munich, Munich, Germany · relAI – Konrad Zuse School of Excellence in Reliable AI, Munich, Germany · German Heart Center Munich, Munich, Germany +3
Department of Control Science and Engineering, College of Electronics and Information Engineering, and Shanghai Institute of Intelligent Science and Technology, Tongji University, Shanghai 200092, China · Department of Electronic Engineering, Faculty of Engineering, The Chinese University of Hong Kong, Hong Kong SAR 999077, China · Shanghai Operation Robot Co., Ltd., Shanghai 201318, China +2