Despite significant advancements in diagnosing and treating psychiatric disorders in recent years, biological tests are still not part of the process. However, researchers using functional MR imaging (fMRI) to examine brain activity are investigating potential biologic markers for diagnosis and treatment of these complex illnesses.
Currently, diagnosing a psychiatric disorder involves several steps, including an evaluation by a physician if symptoms are present. The patient is then referred to a psychiatrist, who—depending on the symptoms and the patient’s behavior—makes a diagnosis based on Diagnostic and Statistical Manual of Mental Disorder (DSM).
“It’s a book of descriptive diagnoses—there’s nothing biological about it,” said Emily Stern, M.D., director of functional and molecular neuroimaging in the Department of Radiology at Brigham and Women’s Hospital (BWH) in Boston. “It’s remarkable that in the 21st Century someone can walk into a doctor’s office and get diagnosed for a psychiatric disorder without one biological test.”
So Dr. Stern, co-director of the Functional Neuroimaging Laboratory at BWH (along with David Silbersweig, M.D., chair of the Department of Psychiatry at BWH), has spent much of her career utilizing functional imaging—particularly fMRI—to examine the brain activity of patients with psychiatric disorders. “The goal is to look at the neurocircuitry that underlies symptom formation across a number of disorders,” she said. “We then try to understand some of the similarities and differences among and between disorders so that we can get a better idea of the biology that causes symptom formation.”
Dr. Stern and colleagues use fMRI to examine a variety of psychiatric illnesses including schizophrenia, affective disorders such as depression and bipolar disease, anxiety disorders, borderline personality disorder, panic disorder and post-traumatic stress disorder. Examining multiple disorders makes sense because many of the symptoms overlap, she said. For example, patients with schizophrenia may become depressed, while those with depression may become anxious and anxious patients often become depressed.
“The descriptive categories in the DSM aren’t really hard and fast,” Dr. Stern said. “Looking at the biological underpinnings of particular symptoms can help us to elucidate a biological framework for understanding all of these disorders.”
In addition to using imaging tools as biomarkers to provide a foundation for possible future diagnosis of psychiatric disorders, one of Dr. Stern’s major goals is to use imaging to guide treatment. For example, it would be extremely helpful to psychiatrists if scanning could help predict treatment response, and there is already some work in the field on this, she said.
“This is so important because some of the very standard forms of pharmacotherapy, such as serotonin reuptake inhibitors (SRIs) can take six to eight weeks to kick in and we wouldn’t know until that point whether the treatment will be efficacious or not,” she said. “That’s a long time to wait for someone who’s potentially suicidal. There is a great need to come up with an effective biological predictor of treatment response.”
She points out that there are already examples of how such work can impact treatment. “One of the first studies we did was to identify certain areas of the brain that underlie hallucination formation in schizophrenia. Investigators at Yale then took this information and used transcranial magnetic stimulation in these regions to reduce hallucinations in schizophrenic patients,” Dr. Stern said. “We’re getting there in very small steps, but we haven’t arrived at the point where we reliably have a biomarker that we know will work in individual patients. That’s our goal.”
Other researchers are using anatomical MR imaging to investigate patterns that emerge from the voxels that represent 3D images—a potential breakthrough for diagnosing and treating mental illness.
At Columbia University, Bradley Peterson, M.D., director of the Center for Developmental Neuropsychiatry, New York, and Ravi Bansal, Ph.D., of the Brain Imaging Laboratory, New York State Psychiatric Institute, New York, have developed an automated method of diagnosing neuropsychiatric illnesses using anatomical MR imaging of the brain.
The spatial variation across voxels and brain subregions, rather than differences within voxels, are mostly likely to represent the abnormalities that produce and define a specific neuropsychiatric illness, according to Drs. Peterson and Bansal.
These patterns across the brain, said Dr. Peterson, are analogous to the dermatomal ridges on a fingertip—a fingerprint. “You don’t identify a person by looking at a single point on that person’s fingertips,” he said. “Instead you look at the overall pattern of those ridges. Rather than analyze the data point by point at every millimeter of the brain, we wanted to capture spatial variations or the spatial pattern of abnormalities, compared with healthy controls or people with other disorders.”
Drs. Peterson and Bansal developed a new algorithm that can diagnose these illnesses based on those spatial variations or patterns. This approach, and the idea that imaging can be used in the diagnosis of psychiatric disorders, has profound implications for patients and their doctors.
Mental health professionals are very good diagnosticians, said Dr. Peterson, and they are usually able to narrow down the possibilities very quickly at the time of initial presentation. But, “only time will most likely tell you what the exact diagnosis turns out to be,” he added.
Just as an internist can order a test to determine blood glucose levels to diagnose Type 2 diabetes, psychiatrists would like to have that same kind of diagnostic certainty to prescribe the best treatment as quickly as possible. Both doctors believe their algorithm can be the lab test that can aid in diagnosis, help treatment planning, put people on the right medications, and give them a more accurate prognosis.
Dr. Peterson also believes the algorithm offers a way of identifying subgroups within disorders, explaining that each should have a different response to specific treatments and different genetic and environmental causes. “We would like to know what those causes and specifically tailored treatments are. And so the prospect of identifying biological subtypes for both research and clinical use is incredibly compelling and exciting, and in my opinion, that’s the greatest potential value of this kind of clinical aid.”
To access the study, “Anatomical Brain Images Alone Can Accurately Diagnose Chronic Neuropsychiatric Illnesses,” by Ravi Bansal, Ph.D., and colleagues in the December 2012 issue of PLOS ONE, go to www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0050698.
For more information on the Functional Neuroimaging Laboratory at Brigham and Women’s Hospital, go to www.functionalneuroimaginglab.org.
Join a global community of leaders in the radiologic sciences.
Continue your education with top-quality learning resources.
With grant applications increasing, the R&E Foundation needs you.
We appreciate your comments and suggestions in our effort to improve your RSNA web experience.
Email Address (required)