The model species of Arabidopsis is well established within the field of botany to study interactions between plants and micro-organisms. Imaging of these interactions is usually performed in light microscopy with diffraction-limited methods. However, these approaches can image either a small field of view at high resolution or a large field of view at low resolution. As such, imaging of bacterial colonies along plant roots cannot be resolved in sufficient numbers for meaningful statistical interpretation. Fourier Ptychographic Microscopy (FPM) combines known properties of Fourier optics with ptychographic imaging and is a low-cost solution for high resolution imaging over large areas. By using a low magnification objective lens and an LED array, FPM varies illumination angles to capture a sequence of images and a reconstruction algorithm combines them into a single high-resolution image. This results in detailed intensity and phase images from original low-resolution datasets. Since its first use, FPM has been primarily focused on improvements to the computational reconstruction and phase retrieval rather than a tool for biological research. We have, for the first time, applied FPM for label-free imaging of Pseudomonas aeruginosa infection of Arabidopsis roots. By training a machine learning network, we have shown that it is possible to identify the pathogen separately from the host. With these new approaches, we aim to identify colonisation patterns to better understand plant-pathogen interactions with our improved resolution over a large field of view.