Answering the first question is straightforward. The complementary anatomical and functional information obtained from separately performed MRI and PET examinations have long been combined either by performing a parallel analysis or by using software co-registration techniques to merge the two data sets. However, one major assumption that has been made in these cases is that no changes in underlying conditions have occurred between the two studies; increasingly, investigators are recognizing that this is not the case, particularly when assessing a subject’s mental state, which may change in a matter of seconds, but also in some diseases such as acute ischemic stroke where changes can occur in a matter of minutes. As we probe illnesses of the mind more thoroughly, the state of the mind and the state of the brain at any given instant become more relevant. This leads to the second and more challenging question. The simultaneous acquisition of PET and MRI data would allow the temporal correlation of these signals. Assessment of a number of diseases could benefit from simultaneous PET/MRI: quantitative measurements of brain cancer, neurodegenerative diseases, acute and chronic stroke, including stroke recovery, breast cancer, pelvic malignancies, cardiac imaging, etc. Identifying which of these many applications will be the first to have widespread clinical impact is an exercise in predicting the future—always a difficult task. Perhaps more importantly, we might more profitably first discuss the fundamental measurement approaches that simultaneous technology opens up. Once implemented and validated, these technical tools could significantly influence a wide range of applications and contribute to the acceptance and successful clinical impact of PET–MRI.

Finally, answering the third question of—why start with brain— is simple: the enormous global burden from neuropsychiatric illnesses is higher than the burden from any other disease category in the developed world. These tremendous unmet medical needs make the brain a logical place to begin investigation. Furthermore, the brain’s favorable imaging characteristics (generally stationary, good size compared with the field of view, multiple intrinsic contrasts of both structure and function/metabolism, etc.) and the wealth of MRI methods that work well in the brain will facilitate these initial efforts. In the following, we first discuss some of the methodological opportunities in a combined scanner and then describe how these could benefit neurological applications, particularly focusing on brain cancer, Alzheimer’s disease, and ischemic stroke.