🤖 AI Summary
Multiphoton microscopy inherently faces a trade-off among phototoxicity, imaging speed, and image quality, resulting in high noise, slow acquisition, and excessive light dosage. To address this, we propose the first deep learning framework that jointly performs denoising and pixel-wise uncertainty estimation—built upon a U-Net variant and Bayesian approximate inference—to enable distribution-free uncertainty quantification directly on real experimental data. We further introduce an adaptive re-sampling paradigm driven by reconstruction uncertainty: only high-uncertainty regions are selectively re-scanned in closed-loop fashion. Validated on human endometrial tissue imaging, our method achieves superior denoising performance over state-of-the-art methods while preserving subcellular structural fidelity. It reduces both acquisition time and total light dose by 120×, delivering simultaneously high fidelity and high reliability. This work establishes an interpretable, verifiable, and intelligent acquisition paradigm for biomedical microscopy.
📝 Abstract
Multiphoton microscopy (MPM) is a powerful imaging tool that has been a critical enabler for live tissue imaging. However, since most multiphoton microscopy platforms rely on point scanning, there is an inherent trade-off between acquisition time, field of view (FOV), phototoxicity, and image quality, often resulting in noisy measurements when fast, large FOV, and/or gentle imaging is needed. Deep learning could be used to denoise multiphoton microscopy measurements, but these algorithms can be prone to hallucination, which can be disastrous for medical and scientific applications. We propose a method to simultaneously denoise and predict pixel-wise uncertainty for multiphoton imaging measurements, improving algorithm trustworthiness and providing statistical guarantees for the deep learning predictions. Furthermore, we propose to leverage this learned, pixel-wise uncertainty to drive an adaptive acquisition technique that rescans only the most uncertain regions of a sample. We demonstrate our method on experimental noisy MPM measurements of human endometrium tissues, showing that we can maintain fine features and outperform other denoising methods while predicting uncertainty at each pixel. Finally, with our adaptive acquisition technique, we demonstrate a 120X reduction in acquisition time and total light dose while successfully recovering fine features in the sample. We are the first to demonstrate distribution-free uncertainty quantification for a denoising task with real experimental data and the first to propose adaptive acquisition based on reconstruction uncertainty
