AI summaryⓘ
The authors address the problem of uneven resolution in 3D microscopy images, where images are clearer in some directions than others. They created SkelEM, a new self-supervised method that separates the process into capturing the basic structure and then adding fine details, which helps avoid blurry or fake features. SkelEM uses a special cycle-consistent technique on sparse slices to improve detail recovery quickly, requiring fewer computational steps. They also introduced a new benchmark dataset, BRAVE-ASR, to test their method's ability to work well across different imaging tools. Their results show improved image quality and better performance in identifying cell membranes compared to other methods.
volume microscopyanisotropic resolutionaxial super-resolutionself-supervised learningdiffusion modelstopological networkcycle-consistencyPlasma-FIBmembrane segmentation
Authors
Bohao Chen, Yanchao Zhang, Yanan Lv, Chenxun Deng, Hua Han, Xi Chen
Abstract
Volume microscopy, including electron and light microscopy, suffers from severe anisotropic resolution due to physical axial sectioning. Existing self-supervised axial super-resolution (ASR) methods face a trilemma bounded by overly smoothed regression textures, structural hallucinations of pure diffusion models, and prohibitive inference latency. In this paper, we propose Skeleton-refinE Microscopy (SkelEM), a self-supervised framework that decouples ASR at the training-signal level: a frozen topological network and a diffusion refiner are optimized by disjoint objectives, separating low-frequency topology formulation from high-frequency detail enhancement. Building on this deterministic skeleton, we exploit a unified cycle-consistent mechanism on input sparse slices to simultaneously extract a real-domain residual prior and bidirectionally align the diffusion refiner, washing away cross-plane artifacts without synthetic bias. By truncating the reverse diffusion process with this physical prior, SkelEM achieves high-fidelity detail restoration in merely $\le 5$ steps. To rigorously assess cross-instrument generalization, we further introduce BRAVE-ASR, a new benchmark of co-aligned anisotropic and isotropic volumes acquired on a Plasma-FIB instrument. Across public benchmarks, SkelEM achieves the most favorable balance across the fidelity-perception trade-off among self-supervised methods, with state-of-the-art downstream membrane segmentation performance and robust zero-shot generalization across distinct modalities.