AnomalousNet: A Hybrid Approach with Attention U-Nets and Change Point Detection for Accurate Characterization of Anomalous Diffusion in Video Data

2025-04-07

AI Summary

Accurate estimation of anomalous diffusion parameters—specifically the anomalous diffusion exponent α and diffusion coefficient D—in short-duration, high-noise videos remains challenging, particularly due to poor identifiability of diffusion-state change points. To address this, we propose the first end-to-end framework integrating an attention-guided U-Net with Bayesian changepoint detection. Our method circumvents conventional reliance on complete particle trajectories and high signal-to-noise ratios by leveraging multi-scale feature reconstruction and spatiotemporal regularization, enabling robust dynamic diffusion modeling under heterogeneous and sparse trajectory conditions. Evaluated on the AnDi-2 video benchmark, our approach achieves a mean absolute error (MAE) < 0.08 for α estimation and a changepoint recall rate > 92%, outperforming state-of-the-art methods. Crucially, it maintains strong robustness under low frame-rate and high-noise regimes, demonstrating practical applicability in real-world microscopy imaging scenarios.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Intelligent Robots: State Estimation
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Content-based information diffusion
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Anomalous diffusion occurs in a wide range of systems, including protein transport within cells, animal movement in complex habitats, pollutant dispersion in groundwater, and nanoparticle motion in synthetic materials. Accurately estimating the anomalous diffusion exponent and the diffusion coefficient from the particle trajectories is essential to distinguish between sub-diffusive, super-diffusive, or normal diffusion regimes. These estimates provide a deeper insight into the underlying dynamics of the system, facilitating the identification of particle behaviors and the detection of changes in diffusion states. However, analyzing short and noisy video data, which often yield incomplete and heterogeneous trajectories, poses a significant challenge for traditional statistical approaches. We introduce a data-driven method that integrates particle tracking, an attention U-Net architecture, and a change-point detection algorithm to address these issues. This approach not only infers the anomalous diffusion parameters with high accuracy but also identifies temporal transitions between different states, even in the presence of noise and limited temporal resolution. Our methodology demonstrated strong performance in the 2nd Anomalous Diffusion (AnDi) Challenge benchmark within the top submissions for video tasks.

Problem

• Estimating anomalous diffusion parameters from noisy video data
• Distinguishing sub-diffusive, super-diffusive, or normal diffusion regimes
• Identifying temporal transitions between diffusion states accurately

Innovation

• Hybrid approach with Attention U-Nets
• Integrates change-point detection algorithm
• Accurate anomalous diffusion parameter inference