Cryo-imaging techniques have been widely used to measure the metabolic state of tissues by capturing reduced nicotinamide adenine dinucleotide (NADH) and flavin adenine dinucleotide (FAD) autofluorescence. However, NADH and FAD fluorescence is sensitive to changes in temperature, which may result in unreliable redox ratio calculations. Here, the relationship between the measured redox ratio and sample surface temperature was analyzed using a standard phantom solution and biological tissues. The results indicated that a temperature < - 100°C was a suitable cryo-imaging temperature window in which redox ratio measuring was immune to temperature fluctuations. These results may serve as a reference for designing and optimizing redox cryo-imaging experiments for quantitatively mapping the metabolic state of biological samples.