MSGID: 1:317/3 641149f1   
   PID: hpt/lnx 1.9.0-cur 2019-01-08   
   TID: hpt/lnx 1.9.0-cur 2019-01-08   
    How neuroimaging can be better utilized to yield diagnostic information   
   about individuals    
      
     Date:   
         March 14, 2023   
     Source:   
         Dartmouth College   
     Summary:   
         Since the development of functional magnetic resonance imaging   
         in the 1990s, the reliance on neuroimaging has skyrocketed as   
         researchers investigate how fMRI data from the brain at rest,   
         and anatomical brain structure itself, can be used to predict   
         individual traits, such as depression, cognitive decline, and brain   
         disorders. But how reliable brain imaging is for detecting traits   
         has been a subject of wide debate.   
      
         Researchers now report that stronger links between brain measures   
         and traits can be obtained when state-of-the-art pattern recognition   
         (or 'machine learning') algorithms are utilized, which can garner   
         high- powered results from moderate sample sizes.   
      
      
         Facebook Twitter Pinterest LinkedIN Email   
   FULL STORY   
   ==========================================================================   
   Since the development of functional magnetic resonance imaging in the   
   1990s, the reliance on neuroimaging has skyrocketed as researchers   
   investigate how fMRI data from the brain at rest, and anatomical brain   
   structure itself, can be used to predict individual traits, such as   
   depression, cognitive decline, and brain disorders.   
      
      
   ==========================================================================   
   Brain imaging has the potential to reveal the neural underpinnings of   
   many traits, from disorders like depression and chronic widespread pain   
   to why one person has a better memory than another, and why some people's   
   memories are resilient as they age. But how reliable brain imaging is   
   for detecting traits has been a subject of wide debate.   
      
   Prior research on brain-wide associated studies (termed 'BWAS') has   
   shown that links between brain function and structure and traits are   
   so weak that thousands of participants are needed to detect replicable   
   effects. Research of this scale requires millions of dollars in investment   
   in each study, limiting which traits and brain disorders can be studied.   
      
   However, according to a new commentary published in Nature, stronger links   
   between brain measures and traits can be obtained when state-of-the-art   
   pattern recognition (or 'machine learning') algorithms are utilized,   
   which can garner high-powered results from moderate sample sizes.   
      
   In their article, researchers from Dartmouth and University Medicine   
   Essen provide a response to an earlier analysis of brain-wide   
   association studies led by Scott Marek at Washington University School   
   of Medicine in St. Louis, Brenden Tervo-Clemmens at Massachusetts General   
   Hospital/Harvard Medical School, and colleagues. The earlier study found   
   very weak associations across a range of traits in several large brain   
   imaging studies, concluding that thousands of participants would be   
   needed to detect these associations.   
      
   The new article explains that the very weak effects found in the earlier   
   paper do not apply to all brain images and all traits, but rather are   
   limited to specific cases. It outlines how fMRI data from hundreds of   
   participants, as opposed to thousands, can be better leveraged to yield   
   important diagnostic information about individuals.   
      
   One key to stronger associations between brain images and traits such   
   as memory and intelligence is the use of state-of-the-art pattern   
   recognition algorithms.   
      
   "Given that there's virtually no mental function performed entirely by   
   one area of the brain, we recommend using pattern recognition to develop   
   models of how multiple brain areas contribute to predicting traits,   
   rather than testing brain areas individually," says senior author Tor   
   Wager, the Diana L. Taylor Distinguished Professor of Psychological and   
   Brain Sciences and director of the Brain Imaging Center at Dartmouth.   
      
   "If models of multiple brain areas working together rather than in   
   isolation are applied, this provides for a much more powerful approach   
   in neuroimaging studies, yielding predictive effects that are four times   
   larger than when testing brain areas in isolation," says lead author   
   Tamas Spisak, head of the Predictive Neuroimaging Lab at the Institute of   
   Diagnostic and Interventional Radiology and Neuroradiology at University   
   Medicine Essen.   
      
   However, not all pattern recognition algorithms are equal and finding the   
   algorithms that work best for specific types of brain imaging data is   
   an active area of research. The earlier paper by Marek, Tervo-Clemmens   
   et al. also tested whether pattern recognition can be used to predict   
   traits from brain images, but Spisak and colleagues found that the   
   algorithm they used is suboptimal.   
      
   When the researchers applied a more powerful algorithm, the effects   
   got even larger and reliable associations could be detected in much   
   smaller samples.   
      
   "When you do the power calculations on how many participants are needed   
   to detect replicable effects, the number drops to below 500 people,"   
   Spisak says.   
      
   "This opens the field to studies of many traits and clinical conditions   
   for which obtaining thousands of patients is not possible, including   
   rare brain disorders," says co-author Ulrike Bingel at University   
   Medicine Essen, who is the head of the University Centre for Pain   
   Medicine. "Identifying markers, including those involving the central   
   nervous system, are urgently needed, as they are critical to improve   
   diagnostics and individually tailored treatment approaches. We   
   need to move towards a personalized medicine approach grounded in   
   neuroscience. The potential for multivariate BWAS to move us towards this   
   goal should not be underestimated."  The team explains that the weak   
   associations found in the earlier analysis, particularly through brain   
   images, were collected while people were simply resting in the scanner,   
   rather than performing tasks. But fMRI can also capture brain activity   
   linked to specific moment-by-moment thoughts and experiences.   
      
   Wager believes that linking brain patterns to these experiences may be a   
   key to understanding and predicting differences among individuals. "One   
   of the challenges associated with using brain imaging to predict traits   
   is that many traits aren't stable or reliable. If we use brain imaging to   
   focus on studying mental states and experiences, such as pain, empathy,   
   and drug craving, the effects can be much larger and more reliable,"   
   says Wager. "The key is finding the right task to capture the state."   
   "For example, showing images of drugs to people with substance use   
   disorders can elicit drug cravings, according to an earlier study   
   revealing a neuromarker for cravings," says Wager.   
      
   "Identifying which approaches to understanding the brain and mind are   
   most likely to succeed is important, as this affects how stakeholders   
   view and ultimately fund translational research in neuroimaging," says   
   Bingel. "Finding the limitations and working together to overcome them   
   is key to developing new ways of diagnosing and caring for patients with   
   brain and mental health disorders."   
       * RELATED_TOPICS   
             o Health_&_Medicine   
                   # Brain_Tumor # Medical_Devices # Nervous_System   
             o Mind_&_Brain   
                   # Brain_Injury # Brain-Computer_Interfaces # Intelligence   
             o Matter_&_Energy   
                   # Medical_Technology # Biochemistry # Ultrasound   
       * RELATED_TERMS   
             o Functional_neuroimaging o Brain_damage o   
             Positron_emission_tomography o Thalamus o   
             Magnetic_resonance_imaging o Alpha_wave o Brain o Amygdala   
      
   ==========================================================================   
   Story Source: Materials provided by Dartmouth_College. Original written   
   by Amy Olson. Note: Content may be edited for style and length.   
      
      
   ==========================================================================   
   Journal Reference:   
      1. Brenden Tervo-Clemmens, Scott Marek, Roselyne J. Chauvin, Andrew   
      N. Van,   
         Benjamin P. Kay, Timothy O. Laumann, Wesley K. Thompson, Thomas E.   
      
         Nichols, B. T. Thomas Yeo, Deanna M. Barch, Beatriz Luna, Damien   
         A. Fair, Nico U. F. Dosenbach. Reply to: Multivariate BWAS can be   
         replicable with moderate sample sizes. Nature, 2023; 615 (7951):   
         E8 DOI: 10.1038/s41586- 023-05746-w   
   ==========================================================================   
      
   Link to news story:   
   https://www.sciencedaily.com/releases/2023/03/230314205337.htm   
      
   --- up 1 year, 2 weeks, 1 day, 10 hours, 50 minutes   
    * Origin: -=> Castle Rock BBS <=- Now Husky HPT Powered! (1:317/3)   
   SEEN-BY: 15/0 106/201 114/705 123/120 153/7715 226/30 227/114 229/110   
   SEEN-BY: 229/111 112 113 307 317 400 426 428 470 664 700 292/854 298/25   
   SEEN-BY: 305/3 317/3 320/219 396/45   
   PATH: 317/3 229/426