The most common way to detect disease is by visual inspection of the suspect tissue. However, the human eye is not optimized for this task because the perceived spectrum of light is divided into three channels, all of which have overlapping spectral sensitivity curves. Here, we present new methods to optimize visually perceived contrast based on spectral differences between normal and abnormal tissue. We apply these methods to the perception of fluorescence emission from the oral cavity. Abnormalities in the oral cavity are optimally perceived when the excitation is between nm. To optimally visualize fluorescence at 340-nm excitation, the emission should be observed through a blue bandpass filter transmitting light at 430 nm.