Investigations into the shape and volume of transtibial prosthetic sockets are complicated because of the difficulty in establishing an accurate reference grid. Magnetic resonance imaging (MRI) presents a possible solution to this problem. However, the reliability of MRI in defining the residual-limb/cast interface depends on the scanned image not being distorted by the materials present. We investigated the potential of MRI technology to establish the desired reference grid. Distortion from the so-called "chemical shift" may influence the MRI when certain materials are used during the casting process. These materials include plaster of paris (POP) and silicone (in the form of an interface liner). POP is commonly used to capture the shape of the residual limb. However, if the casting technique requires the use of a silicone liner, the liner is placed over the residual limb first and then the POP is applied over the liner. Experimental results indicate that the materials used do not distort or interfere with the scanned image. The object segmentation process that extracts the bone and skin from an MRI scan and enables the establishment of the required reference grid was explored. Results show that extracting the bone structure and using it as the reference grid to quantify the differences in volume and shape of the soft tissues of the residual limb is feasible.