In recent decades, we have witnessed the importance of medical imaging, for example computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomography (PET), mammography, ultrasound, X-rays. Detection, diagnosis and treatment. of diseases that are important for early medicine. In clinics, interpretation of medical images is performed primarily by human experts (such as radiologists and physicians). However, due to dramatic changes in pathology and possible fatigue for human experts, researchers and clinicians have recently begun to benefit from computer-assisted interventions. Although compared to the advancement of medical imaging technology, its advancement in computational analysis of medical images is late, but recently it has been continuously improving with the help of machine learning techniques.
According to a study published in “Biological Psychiatry: Cognitive Neuroscience and Neuroimaging,” medical imaging technology has helped researchers discover a brain biomarker that may increase the risk of post-traumatic stress disorder (PTSD) after a significant brain injury.
PTSD is a complex mental illness caused by physical and psychological trauma. Symptoms can include depression, anxiety, and cognitive decline. How these symptoms occur is not fully understood or is unpredictable. If doctors can better predict who will develop PTSD, they can improve treatment methods and treatment effects. With the help of new machine learning and medical imaging techniques that seems to be an approachable goal.
Dr. Murray Stein, Distinguished Professor of Psychiatry and Family Medicine and Public Health, at the University of California, San Diego, said: “In recent years, the relationship between TBI ( traumatic brain injury) and PTSD has been increasingly affected due to the fact which the research has shown, that there is considerable overlap in the risk factors and symptoms. .”
“In this study, we were able to use data from TRACK-TBI, a large longitudinal study of patients who were present (at some point) in the Emergency Department with TBIs serious enough to warrant CT (computed tomography) scans.”
Researchers followed over 400 TBI patients, assessing them at three and six months after their brain injury. At three months, 77 participants — or 18 percent — had likely PTSD. At six months, 70 participants, or 16 percent, did. All subjects underwent brain injury after injury.
Stein said: “MRI studies performed within two weeks of injury are used to measure the data of key brain structures believed to be related to PTSD.” “We found that the data of some of these structures can predict PTSD three months after injury.”
Specifically, within three months, the smallest volume of the brain area called the cingulate cortex, the frontal epithelium, and the insula can predict PTSD. These areas are related to arousal, attention, and emotional regulation. At six months, structural imaging did not predict PTSD.
The researchers noted that these results are consistent with previous studies. Previous studies have shown that several of these brain regions in PTSD patients are smaller in size, and studies have shown that reduction in cortical volume may be a risk factor for the development of PTSD.
Together, the findings suggest that a brain reserve, or higher cortical volumes, may provide some resilience against PTSD.
Biomarkers (suggested by the AI) of brain volume differences are not sufficient to provide clinical guidance, but the research team noted that these findings may provide a basis for future research.
“The findings do pave the way for future studies to look even more closely at how these brain regions may contribute to (or protect against) mental health problems such as PTSD,” Stein said.
The results of the study demonstrate the ability of medical imaging to help improve PTSD diagnosis and treatment.
“This very important study uses MRI imaging to zoom in on why some people develop PTSD after trauma, while others do not,” said , Camer Len Carter (MD), editor of Bio-psychiatry: Cognitive Neuroscience and Neuroimaging.
“It also lays the groundwork for future research aimed at using brain imaging to help predict that a person is at increased risk and may benefit from targeted interventions to reduce the clinical impact of a traumatic event.”
This research adds to recent efforts of healthcare and technology to improve the diagnosis of PTSD. In October 2020, researchers at the Boston University School of Public Health (BUSPH) used machine learning to reduce 6 out of 20 problems in diagnosing PTSD, while maintaining the accuracy of the elderly population.
“We found that some PTSD projects can be removed because they did not make a substantial contribution to the accurate prediction of PTSD compared to other PTSD projects. It is likely that some of these items can be removed because they are redundant with other items. Other items may be mobile because they are not specific enough for PTSD.