Pathogen infections are among the main factors that threaten crop production and quality worldwide, being their early detection a step of paramount importance to efficiently manage plant pathologies. The methods currently employed for plant disease detection often require the presence of visible signs of the infection, which affects the efficacy of protection measures, since symptoms only appear later on. Hence, alternative methods based on proximal optical sensing (POS) have recently been investigated, introducing new perspectives in the phytopathology field. They rely on findings that interactions between host plants and pathogens induce changes in the biochemical and internal structure of leaves, causing modifications in their optical properties. Within this scope, a study carried out in a walk-in plant growth chamber, analyzed the potential of POS as an effective approach for early disease detection. A compact, modular sensing system, combining direct UV-Vis spectroscopy with optical fibers, supported by a robust Self-Learning Artificial Intelligence (SLAI), was applied to evaluate the modifications promoted by the bacteria Xanthomonas euvesicatoria (Xeu) in tomato leaves (cv. Cherry). Plant infection was achieved by spraying a bacterial suspension (10⁸ CFU mL⁻¹) until run-off occurred, and a similar approach was followed for the control group where only water was applied. A total of 270 spectral measurements – from 195 to 1100 nm – were performed on leaves, on five different time instances, including pre- and post-inoculation measurements. The acquired data from the assays performed, from both healthy and inoculated leaves, was then analyzed by an innovative SLAI, which allowed their distinction and differentiation. These results suggest that this in vivo, non-destructive POS technique may be promising for assessing changes in the spectral behavior of diseased crop leaves. Consequently, the present work encourages future usage and improvement of these optical devices, as well as its enhancement to perform diagnosis directly in the field.