[论文解读] λSplit: Self-Supervised Content-Aware Spectral Unmixing for Fluorescence Microscopy

Federico Carrara, Talley J. Lambert+2 more|arXiv (Cornell University)|Mar 24, 2026

Cell Image Analysis Techniques被引用 0

一句话总结

λSplit 是一个基于物理信息的自监督深度生成模型，具有一个可微分的谱混合器，学习分层先验以从光谱显微数据中分离荧光团浓度，具鲁棒性对噪声与光谱重叠。

ABSTRACT

In fluorescence microscopy, spectral unmixing aims to recover individual fluorophore concentrations from spectral images that capture mixed fluorophore emissions. Since classical methods operate pixel-wise and rely on least-squares fitting, their performance degrades with increasingly overlapping emission spectra and higher levels of noise, suggesting that a data-driven approach that can learn and utilize a structural prior might lead to improved results. Learning-based approaches for spectral imaging do exist, but they are either not optimized for microscopy data or are developed for very specific cases that are not applicable to fluorescence microscopy settings. To address this, we propose λSplit, a physics-informed deep generative model that learns a conditional distribution over concentration maps using a hierarchical Variational Autoencoder. A fully differentiable Spectral Mixer enforces consistency with the image formation process, while the learned structural priors enable state-of-the-art unmixing and implicit noise removal. We demonstrate λSplit on 3 real-world datasets that we synthetically cast into a total of 66 challenging spectral unmixing benchmarks. We compare our results against a total of 10 baseline methods, including classical methods and a range of learning-based methods. Our results consistently show competitive performance and improved robustness in high noise regimes, when spectra overlap considerably, or when the spectral dimensionality is lowered, making λSplit a new state-of-the-art for spectral unmixing of fluorescent microscopy data. Importantly, λSplit is compatible with spectral data produced by standard confocal microscopes, enabling immediate adoption without specialized hardware modifications.

研究动机与目标

• 在光谱重叠与噪声存在下，说明在荧光显微成像中进行鲁棒光谱解混的需求。
• 开发一个自监督、物理信息驱动的生成模型，为未混合的荧光团图提供结构先验。
• 引入可微谱混合器以强化与成像过程的一致性，并实现端到端训练。
• 在多样化采集设置下通过合成基准和真实数据集展示鲁棒性与先进性能。

提出的方法

• 一个 Ladder Variational Autoencoder 骨架学习荧光团浓度地图的分层潜在表示。
• 一个完全可微的 Spectral Mixer 实现线性成像模型 S = M U，将未混合的浓度 U 映射到光谱空间，使用离散化的荧光谱。
• 训练最大化带有光谱重建项的条件 LVAE 目标：L = E_{q(z|S)}[L_spMSE(S, S_hat(z))] + KL(q(z|S) || p(z))。
• 谱混合器通过用固定混合矩阵 M 从预测的 U 重构 S，从而强制物理一致性。
• 发射光谱 M 归一化以保持总强度，并来自 FP 数据库或测量值。
• 模型在 2D/3D 中使用 LVAE，采样 50 个后验预测用于 MMSE 估计；在 patches 上用 Adamax 训练，支持提前停止与混合精度。

Figure 1: Spectral Imaging and Unmixing in a nutshell. A biological sample is labeled with $F$ fluorescent probes (FPs) whose emission spectra typically overlap. While for multiplexed imaging each FP is imaged in isolation using $F$ appropriate band-pass filters (see Multiplexed Imaging Channels), s

实验结果

研究问题

• RQ1 一个自监督、物理信息驱动的深度生成模型是否可以在高光谱重叠与噪声下超越传统与基于学习的解混方法？
• RQ2 引入可微谱混合步骤如何影响在不同光谱维度下的重建保真度与鲁棒性？
• RQ3 在仅有光谱数据且缺乏 ground-truth 未混合图的情况下，学习的结构先验在多大程度上改善未混合荧光团图？
• RQ4 该方法是否可在不改变硬件的情况下与标准共聚焦显微镜数据兼容？

主要发现

• λSplit 在受控基准测试中在不同噪声、光谱重叠和光谱维度下持续超越基线方法。
• 在高噪声与强光谱重叠条件下，λSplit 显示出更强的鲁棒性与降噪能力，使未混合图像的 SNR 提高。
• 该方法在利用学习先验的同时，保持与最先进的 PSNR、MS-SSIM 指标及感知指标的竞争力。
• 与监督基线相比，λSplit 在参数量极小（约 3M 参数）且无需 ground-truth 未混合图的情况下实现了相似或更好性能。
• 在减少光谱条带数量时，λSplit 展现稳定性，在欠定情形（L < F）下优于经典方法。
• 定性结果显示边界更清晰，结构细节保留更好，得益于隐式去噪与结构先验。

Figure 2: Proposed architecture of $\lambda$ Split. The model builds on an LVAE backbone [ 35 ] , where a bottom-up encoder produces features $h_{i}$ at multiple hierarchy levels. At the highest hierarchy level we employ a default multivariate Gaussian prior, followed by learnable top-down priors $p