The thalamus is an important relay structure in the brain that may be relevant to a variety of brain diseases. It is divided into multiple subnuclei with different cortical connections. The medial dorsal (MD) nucleus is particularly important because it forms key connections with the prefrontal cortex. The current study reports precise and efficient methods for measuring the whole thalamus and the MD with MRI that have a high degree of interrater reliability. A multispectral image acquisition and novel image processing technique were used to improve structure visibility. The tricolor image assigns a color to each of the T1, T2, and PD weighted images, represented by red, green, and blue, respectively. The manually defined regions were then used to train an artificial neural network (ANN) to automatically define both the whole thalamus and the MD. The ANN provides an efficient automated method, making studies using larger sample sizes more feasible.