Filament tracing in noisy microscopy images using topology guided evolution
FPicker: Topology-Guided Evolution for Filament Tracing in Low-SNR Microscopy
Computer Vision and Pattern RecognitionArtificial Intelligence
Summary
Tracing tiny, threadlike structures in very noisy microscope images is difficult because the usual software tools either break these threads apart or get confused by overlaps. The authors developed a new method called FPicker that uses the shape and connection rules of these threads to track them better. Their approach works well even for very noisy images and can recognize the intrinsic shape instead of just pixel details. It performs effectively on both simulated and real microscope data, suggesting it can help scientists study microscopes images with poor signal quality.
What this means in practice
- •For microscopy image analysts: Automatically trace filament structures in extremely noisy cryo-EM images to improve 3D reconstructions.
- •For scientific software developers: Integrate topology-guided filament tracing tools into imaging pipelines for better noise resilience in bioimaging applications.
Authors
Tingyin Zhao, Mingtao Huang, Yuan Shen
Abstract
Automating filament tracing in Cryo-Electron Microscopy (Cryo-EM) is essential for 3D helical reconstruction but challenged by intersecting topologies and extremely low Signal-to-Noise Ratios ($\text{SNR} = σ_s^2/σ_n^2$ < 0.1 or -10 dB). Existing paradigms fail: pixel-wise segmenters suffer from severe topological fracturing, box-based detectors face ghost center drift, sequential trackers derail due to error accumulation, and traditional active contours collapse under artificial closed-curve constraints. To resolve these bottlenecks, we present FPicker, the first topology-guided framework reconciling these incompatibilities. It unifies perception via a center-endpoint representation and an open-curve evolution module to explicitly model non-cyclic connectivity. On simulated benchmarks, FPicker outperforms top baselines by over $40\%$ relative gain in mean spatio-angular precision (mSAP) and reduces topological gap rates by over $60\%$ under extreme noise ($-20\text{ dB}$). By learning intrinsic physical geometry rather than local texture, FPicker demonstrates strong potential as a resilient geometric backbone. Its zero-shot performance on the real-world EMPIAR dataset exhibits robust topological resistance, achieving a state-of-the-art 82.9\% mSAP upon fine-tuning. Our results also suggest modeling physical priors is a highly robust path toward bridging the sim-to-real gap in signal-starved scientific imaging. The code is publicly available at: https://github.com/tomzhaosky/FPicker.