🤖 AI Summary
Deep-tissue microscopy imaging suffers from severe contrast degradation due to optical scattering, and the absence of ground-truth (GT) high-quality images hinders supervised deep learning approaches. To address this, we propose a GT-free deep contrast enhancement framework. First, we formulate an interpretable, physics-inspired degradation model to synthesize high-fidelity training data. Second, we design an invertible neural network coupled with an out-of-distribution generalization mechanism to ensure robust transfer to real-world degradation distributions. Third, we introduce a controllable iterative prediction scheme that dynamically balances contrast enhancement and structural fidelity. Experimental results demonstrate that our method achieves a 2.3× average contrast improvement under GT-free conditions, while preserving key cellular structures at analyzable resolution—enabling reliable downstream quantitative analysis.
📝 Abstract
Microscopy images are crucial for life science research, allowing detailed inspection and characterization of cellular and tissue-level structures and functions. However, microscopy data are unavoidably affected by image degradations, such as noise, blur, or others. Many such degradations also contribute to a loss of image contrast, which becomes especially pronounced in deeper regions of thick samples. Today, best performing methods to increase the quality of images are based on Deep Learning approaches, which typically require ground truth (GT) data during training. Our inability to counteract blurring and contrast loss when imaging deep into samples prevents the acquisition of such clean GT data. The fact that the forward process of blurring and contrast loss deep into tissue can be mod-eled, allowed us to propose a new method that can circumvent the problem of unobtainable GT data. To this end, we first synthetically degraded the quality of microscopy images even further by using an approximate forward model for deep tissue image degradations. Then we trained a neural network that learned the inverse of this degradation function from our generated pairs of raw and degraded images. We demonstrated that networks trained in this way can be used out-of-distribution (OOD) to improve the quality of less severely degraded images, e.g. the raw data imaged in a microscope. Since the absolute level of degradation in such microscopy images can be stronger than the additional degradation introduced by our forward model, we also explored the effect of iterative predictions. Here, we observed that in each iteration the measured image contrast kept improving while detailed structures in the images got increasingly removed. Therefore, dependent on the desired downstream analysis, a balance between contrast improvement and retention of image details has to be found.
