Author name cluster

Lei Fan

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

6 papers

2 author rows

AAAI Conference 2026 Conference Paper

Flexible Concept Bottleneck Model

Xingbo Du
Qiantong Dou
Lei Fan
Rui Zhang

Concept bottleneck models (CBMs) improve neural network interpretability by introducing an intermediate layer that maps human-understandable concepts to predictions. Recent work has explored the use of vision-language models (VLMs) to automate concept selection and annotation. However, existing VLM-based CBMs typically require full model retraining when new concepts are involved, which limits their adaptability and flexibility in real-world scenarios, especially considering the rapid evolution of vision-language foundation models. To address these issues, we propose Flexible Concept Bottleneck Model (FCBM), which supports dynamic concept adaptation, including complete replacement of the original concept set. Specifically, we design a hypernetwork that generates prediction weights based on concept embeddings, allowing seamless integration of new concepts without retraining the entire model. In addition, we introduce a modified sparsemax module with a learnable temperature parameter that dynamically selects the most relevant concepts, enabling the model to focus on the most informative features. Extensive experiments on five public benchmarks demonstrate that our method achieves accuracy comparable to state-of-the-art baselines with a similar number of effective concepts. Moreover, the model generalizes well to unseen concepts with just a single epoch of fine-tuning, demonstrating its strong adaptability and flexibility.

PDF Details DOI

EAAI Journal 2026 Journal Article

Review of physics-informed neural networks in hemodynamics

Xianglong Yu
Yu Hu
Rui Guo
Lei Fan
Haiyan Ding
Jingjing Xiao

The circulatory system sustains physiological function through oxygen transport, nutrient delivery, and waste clearance, all of which rely on efficient blood flow. Accurate characterization and quantification of hemodynamics are essential for the diagnosis and treatment of cardiovascular diseases. However, assessing blood flow in a noninvasive and real-time manner remains a major challenge, as current imaging modalities often suffer from limited spatial and temporal resolution, while traditional computational fluid dynamics algorithms are computationally intensive and sensitive to anatomical and physiological uncertainties. Physics-informed neural networks (PINNs), combining physical laws with data-driven learning, provide a promising framework to connect computational modeling with clinical applications. In this review, we provide a comprehensive overview of recent advances in the application of PINNs to hemodynamics. We introduce theoretical foundations, highlight methodological innovations, and discuss applications in simulating blood flow under physiological and pathological conditions, as well as in estimating clinically relevant hemodynamic parameters. Importantly, our analysis highlights that PINNs achieve comparable accuracy to traditional methods while unlocking novel opportunities for patient-specific diagnosis and risk prediction. We conclude with a discussion of the benefits, current limitations, and future directions of PINNs in cardiovascular research, underscoring the transformative potential to accelerate clinical translation through interdisciplinary collaboration.

Details DOI

ICRA Conference 2025 Conference Paper

Distributed Perception Aware Safe Leader Follower System via Control Barrier Methods

Richie R. Suganda
Tony Tran
Miao Pan
Lei Fan
Qin Lin
Bin Hu

This paper addresses a distributed leader-follower formation control problem for a group of agents, each using a body-fixed camera with a limited field of view (FOV) for state estimation. The main challenge arises from the need to coordinate the agents' movements with their cameras' FOV to maintain visibility of the leader for accurate and reliable state estimation. To address this challenge, we propose a novel perception-aware distributed leader-follower safe control scheme that incorporates FOV limits as state constraints. A Control Barrier Function (CBF) based quadratic program is employed to ensure the forward invariance of a safety set defined by these constraints. Furthermore, new neural network based and double bounding boxes based estimators, combined with temporal filters, are developed to estimate system states directly from real-time image data, providing consistent performance across various environments. Comparison results in the Gazebo simulator demonstrate the effectiveness and robustness of the proposed framework in two distinct environments.

Details

AAAI Conference 2025 Conference Paper

GRPose: Learning Graph Relations for Human Image Generation with Pose Priors

Xiangchen Yin
Donglin Di
Lei Fan
Hao Li
Wei Chen
Gouxiaofei
Yang Song
Xiao Sun

Recent methods using diffusion models have made significant progress in human image generation with various control signals such as pose priors. However, existing efforts are still struggling to generate high-quality images with consistent pose alignment, resulting in unsatisfactory output. In this paper, we propose a framework that delves into the graph relations of pose priors to provide control information for human image generation. The main idea is to establish a graph topological structure between the pose priors and latent representation of diffusion models to capture the intrinsic associations between different pose parts. A Progressive Graph Integrator (PGI) is designed to learn the spatial relationships of the pose priors with the graph structure, adopting a hierarchical strategy within an Adapter to gradually propagate information across different pose parts. Besides, a pose perception loss is introduced based on a pretrained pose estimation network to minimize the pose differences. Extensive qualitative and quantitative experiments conducted on the Human-Art and LAION-Human datasets clearly demonstrate that our model can achieve significant performance improvement over the latest benchmark models.

PDF Details DOI

IJCAI Conference 2025 Conference Paper

Multimodal Cancer Survival Analysis via Hypergraph Learning with Cross-Modality Rebalance

Mingcheng Qu
Guang Yang
Donglin Di
Tonghua Su
Yue Gao
Yang Song
Lei Fan

Multimodal pathology-genomic analysis has become increasingly prominent in cancer survival prediction. However, existing studies mainly utilize multi-instance learning to aggregate patch-level features, neglecting the information loss of contextual and hierarchical details within pathology images. Furthermore, the disparity in data granularity and dimensionality between pathology and genomics leads to a significant modality imbalance. The high spatial resolution inherent in pathology data renders it a dominant role while overshadowing genomics in multimodal integration. In this paper, we propose a multimodal survival prediction framework that incorporates hypergraph learning to effectively capture both contextual and hierarchical details from pathology images. Moreover, it employs a modality rebalance mechanism and an interactive alignment fusion strategy to dynamically reweight the contributions of the two modalities, thereby mitigating the pathology-genomics imbalance. Quantitative and qualitative experiments are conducted on five TCGA datasets, demonstrating that our model outperforms advanced methods by over 3. 4% in C-Index performance. Code: https: //github. com/MCPathology/MRePath.

PDF Details DOI

AIIM Journal 2024 Journal Article

Rapid estimation of left ventricular contractility with a physics-informed neural network inverse modeling approach

Ehsan Naghavi
Haifeng Wang
Lei Fan
Jenny S. Choy
Ghassan Kassab
Seungik Baek
Lik-Chuan Lee

Physics-based computer models based on numerical solutions of the governing equations generally cannot make rapid predictions, which in turn limits their applications in the clinic. To address this issue, we developed a physics-informed neural network (PINN) model that encodes the physics of a closed-loop blood circulation system embedding a left ventricle (LV). The PINN model is trained to satisfy a system of ordinary differential equations (ODEs) associated with a lumped parameter description of the circulatory system. The model predictions have a maximum error of less than 5% when compared to those obtained by solving the ODEs numerically. An inverse modeling approach using the PINN model is also developed to rapidly estimate model parameters (in ∼ 3 min) from single-beat LV pressure and volume waveforms. Using synthetic LV pressure and volume waveforms generated by the PINN model with different model parameter values, we show that the inverse modeling approach can recover the corresponding ground truth values for LV contractility indexed by the end-systolic elastance E e s with a 1% error, which suggests that this parameter is unique. The estimated E e s is about 58% to 284% higher for the data associated with dobutamine compared to those without, which implies that this approach can be used to estimate LV contractility using single-beat measurements. The PINN inverse modeling can potentially be used in the clinic to simultaneously estimate LV contractility and other physiological parameters from single-beat measurements.

Details DOI