The Difference between an MRI Research Finding and a Psychiatric Diagnosis — BrainPost | Easy-to-read summaries of the latest neuroscience publications

“Why won’t my doctor…?”

Diagnosing a psychiatric illness is not always straightforward. Let’s take depression for example. The symptoms are not visible to the naked eye and can vary from patient to patient. On the other hand, there seems to be a wealth of brain imaging studies showing differences between people with and without depression. With all of these studies, it is tempting to think, “why won’t my doctor just give me an MRI scan to see if I have depression?”

To understand why, let’s take a simplified example: we have two groups of people, one group of males and another group of females (setting aside the complexities of gender identity for the moment). The only thing we know about these people is their height. Unless we have a very strange sample, we expect the groups to reflect the general population, with the female group having the smaller average height, and the male group having the larger average height. This, in our analogy, is similar to a research finding. Now, what if I pick a person at random and tell you that their height is 175 cm (5 feet 7 inches)? Could you reliably tell if they are from the male or female group? Not at all. This is similar to the challenge that arises when diagnosing a single person. To summarize, researchers can find subtle differences when they compare large groups of people with different psychological conditions, and this helps us to understand these conditions better. However, it can be difficult to classify any one person based on a brain scan.

Added to this is the expense of an MRI scan—these can cost hundreds to thousands of dollars an hour, and scheduling one will take time. Your doctor is constantly engaged in a cost-benefit analysis, trying to get you the most reliable diagnosis in the shortest time, and oftentimes an MRI may not be worth the cost. Why pay hundreds of dollars for a brain scan when a carefully validated questionnaire would also be effective?

So why are MRIs useful?

Firstly, there are several neurological conditions that can be diagnosed with an MRI, including strokes, tumors, and multiple sclerosis. The brain differences here are much easier to pick out, assuming sufficient training, and an MRI can help to determine definitively whether or not someone has the disease. In psychiatric conditions, MRI research has led to discovering much about the mechanisms behind different conditions. Let’s return to the previous example of depression. MRI has helped researchers to understand the involvement of certain brain regions in depression, like the frontal lobe and the amygdala, including how these regions differ in terms of their structure, function, and connectivity with other regions. This is also true for many other psychiatric conditions, such as obsessive-compulsive disorder, anxiety disorders or schizophrenia. In addition, MRI technology and analysis techniques are becoming more advanced every day. Researchers are now developing new MRI methods that may be able to visualize things at a higher resolution that we couldn’t see before. Techniques are also being developed that will help us to look at individual brain differences, and this can guide various personalized treatment approaches in psychiatry. Many also hope to employ artificial intelligence to identify more subtle abnormalities in scans and find people who might benefit from preventative treatments. If MRI costs were brought down somehow, the landscape of diagnosis might change dramatically as well.

What’s the bottom line?

What ultimately matters for a diagnosis is not always what your brain looks like, but rather what symptoms you’re having and how your daily functioning is affected. Although there are some diseases where MRI can be used to diagnose a patient, there are many cases where an MRI is complimentary or not necessarily needed. The usefulness of MRI in treatment will depend on whether looking at an MRI can help a clinician decide between various treatments (and whether it is worth the time and expense of a scan). MRI has provided immense value in understanding the causes and progression of many psychiatric diseases and this is crucial for the development of future treatments. As technology continues to advance, and if costs lower over time, MRI may become even more applicable to a wide variety of uses like diagnosis, guiding treatment, and monitoring recovery.

References

Zhang, F. F., Peng, W., Sweeney, J. A., Jia, Z. Y., & Gong, Q. Y. (2018). Brain structure alterations in depression: Psychoradiological evidence. CNS neuroscience & therapeutics, 24(11), 994–1003. https://doi.org/10.1111/cns.12835

Lo, A., Chernoff, H., Zheng, T., & Lo, S. H. (2015). Why significant variables aren't automatically good predictors. Proceedings of the National Academy of Sciences of the United States of America, 112(45), 13892–13897. https://doi.org/10.1073/pnas.1518285112

Hunter SF. Overview and diagnosis of multiple sclerosis. Am J Manag Care. 2016 Jun;22(6 Suppl):s141-50. PMID: 27356023.

Foland-Ross, L. C., Sacchet, M. D., Prasad, G., Gilbert, B., Thompson, P. M., & Gotlib, I. H. (2015). Cortical thickness predicts the first onset of major depression in adolescence. International journal of developmental neuroscience : the official journal of the International Society for Developmental Neuroscience, 46, 125–131. https://doi.org/10.1016/j.ijdevneu.2015.07.007

Jollans, L., Boyle, R., Artiges, E., Banaschewski, T., Desrivières, S., Grigis, A., Martinot, J. L., Paus, T., Smolka, M. N., Walter, H., Schumann, G., Garavan, H., & Whelan, R. (2019). Quantifying performance of machine learning methods for neuroimaging data. NeuroImage, 199, 351–365. https://doi.org/10.1016/j.neuroimage.2019.05.082