Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more ➡
Standard view
Full view
of .
Add note
Save to My Library
Sync to mobile
Look up keyword
Like this
0 of .
Results for:
No results containing your search query
P. 1
Diagnosis by Default

Diagnosis by Default

Ratings: (0)|Views: 470|Likes:
Published by Angel Martorell
Article from Nature Medicine 18, 338–340 (2012)
Article from Nature Medicine 18, 338–340 (2012)

More info:

Published by: Angel Martorell on Mar 14, 2012
Copyright:Attribution Non-commercial


Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See More
See less





Signal to diagnoise
Scientists performed the first human MRIexamination in the late 1970s, but it tookabout fifteen years for them to use thesame technology to track brain signals withfMRI. In the early days of fMRI, the expertstrying to analyze the data had a tough timediscerning the ‘signals’ of altered brainactivity from the ‘noise’ of regular, run-of-the-mill blood flow that sustains the organ.Neuroscientists hoping to tease this apartwould compare fMRI scans of volunteerscompleting tasks with those of the samepeople when idle. The hope was to deducethe function of various brain regions by canceling out the background noise emittedby the brain during rest.A twist came in 1995, when Hyde and hisgraduate student at the time, Bharat Biswal,became curious as to whether resting-statescans had distinct characteristics of their own.The Wisconsin researchers had noticedthat the noise in the fMRI data seemed to beunusually high, and they wondered whetherthe racket was coming from the brain itself.reveal more intimate details about how brainregions communicate or fall out of sync as aresult of disease. The approach, which mighteventually be used as a diagnostic tool, has evencaught the attention of some big-name drugdevelopers.“There’s tremendous interest in the intrinsicactivity in the brain,” says Marcus Raichle, anearly pioneer in the field at the WashingtonUniversity School of Medicine in St. Louis. Thegrowth in research into the brain’s resting statehas been “exponential,” he adds.Already, scientists have found inklingsof differences in the resting-state brainactivity of people with Alzheimer’sdisease and schizophrenia, among otherneurological disorders. And only a year ago,a competition was launched to find the bestanalytical approach to detect attention deficithyperactivity disorder (ADHD) from resting-state scans.“We’re on an amazing adventure. It really is,” says James Hyde at the Medical College of Wisconsin in Milwaukee, who first publishedon the subject nearly two decades ago.Even though the space inside a brain scanneris snug, volunteers in imaging studies usually have their hands full. Lying inside the machine,surrounded by hulking magnets, they might beasked to tap their fingers, listen to sounds orrecognize faces flashing on a screen. Whilethe subjects respond to these prompts, pulsesof radio waves travel through their heads,knocking the protons in the brain out of order so that the machine can measure theirresonance. Thankfully for researchers, oxygen-rich blood has a different magnetic resonancethan oxygen-depleted blood, so they can detect variations in blood flow—and in brain activity by proxy.Such functional magnetic resonanceimaging (fMRI) studies have offered insightson everything from how people respond toadvertising to how they process language.But, increasingly, a group of neuroscientistsbelieve that scientists focusing on task-basedbrain scans are missing the big picture. Overthe last decade, momentum has swelled aroundthe idea that examining resting-state fMRI,captured as people lie idle in the scanner, could
Brain scans that map differences in how brain regions communicate while people lie idle in the imaging machine couldone day provide clues about afflictions ranging from Alzheimer’s disease to attention disorders.
Roxanne Khamsi
findsout why these so-called ‘resting state’ scans have made researchers and drug companies sit up and take notice.
volume 18|number 3|march 2012
nature medicine
   L  u  c  a  s   G  r  a  y  a  n   d   K  r   i  s   t   i  n   S   l  a  v   i  n
news feature
nature medicine
volume 18|number 3|march 2012
Together, these findings hint at a futurewhen functional connectivity of brain regionscould serve as a better early indicator of Alzheimer’s than memory tests or measuresof certain brain proteins.
Making the connection
As more teams attempt to obtain resting-statescans from patients and healthy volunteers,the research community has sought a way to bring the information all together. InDecember 2009, investigators from 35 centersaround the world pooled resting-state fMRIrecords from more the 1,400 healthy subjectsand released their data online through whatbecame known as the 1,000 FunctionalConnectomes Project
. According to MichaelMilham, founding director of the Centerfor the Developing Brain at the Child MindInstitute in New York and a co-creator of theproject, the resource provided a repository of scans for people to evaluate using novelalgorithms in hopes of finding improveddiagnostic analyses.But the project, as initially conceived, waslimited in that each brain scan was taggedwith only the age and sex of the individual,but not any clinical information. Also, theimaging speed of the fMRI machines used varied across sites, making it difficult tocompare the data sets. So, Milham andhis collaborators around the globe havemorphed the effort to address these issuesinto what’s known as the InternationalNeuroimaging Data-Sharing Initiative, whichhas scans tagged with diagnostic backgroundinformation and includes both prospectiveand retrospective data.Meanwhile, in 2010 the US NationalInstitutes of Health funded sites for itsdisparate regions of the default network—including the medial prefrontal cortex atthe front of the brain and the lateral parietalcortex on the side—get going in concert whenpeople are idle. “That’s part of the mystique of the default network,” Greicius says. “No onereally knows what it does.”To better understand the healthy defaultnetwork, researchers turnedto studying faulty baselinestates in people with variousneurological and behavioraldisorders, ranging fromepilepsy and coma todepression and Parkinson’sdisease. One of the earlieststudies of this kind, led by Mark Lowe, director of theHigh-Field MRI branch atthe Cleveland Clinic Foundation in Ohio,analyzed finger-tapping and resting-statebrain scans in 20 people with multiplesclerosis and 16 healthy controls andfound that, compared with their controlcounterparts, individuals with the diseasehad less synchronicity in the ‘sensorimotorregions’ of the brain, which are involved inmovement control and sensory perception
.Investigation into the default-modenetwork of people with Alzheimer’s diseasehas also revealed intriguing results. Studieshave found decreased coordinated activity between the hippocampus—a memory center—and other brain regions during theresting state
. And an experiment publisheda few years ago even found a link betweena variant in the
gene that raises therisk of Alzheimer’s and unusual activity inthe default-mode network compared withcontrols
.“I went around screaming about the noise ineveryone’s fMRI scans,” says Hyde. “I gave[Biswal] the assignment and I said ‘The noiseis too high.’”For his PhD dissertation, Biswal looked atfMRI scans of eleven subjects who lay at restfor five minutes and later did a finger-tappingtask for five minutes. After filtering outchanges in blood flow causedby the participants’ heart beatsand breathing, the analysisrevealed a sustained level of activity in the motor cortexof people at rest, indicatingthat the noise was indeedphysiological and neurologicalin basis
. “I saw that therewas still this low-frequency signal that was present in thebrain at resting state,” says Biswal, now at theUniversity of Medicine and Dentistry of NewJersey in Newark. “Therefore, the questionwas, ‘What was its role?’”According to Biswal, who trained as anelectrical engineer before joining Hyde’slab,
neuroscientists had simply failed toappreciate that the circuitry of the restingbrain functioned in a meaningful pattern,much like a computer in standby mode stillhas purposefully activated circuits.Biswal’s quest to understand whetherthe activity in the resting brain can have value came at a time when the imaging fieldin general still had not found its footing.“Back in 1995, people were still testingwhether functional MRI was of any use,”says Alexandre Coimbra, a senior clinicalimaging scientist at Genentech, a South SanFrancisco–based subsidiary of the Swiss druggiant Roche. “It was very far-fetched at thetime what he was proposing.”The idea started to pick up steam two yearslater when Raichle and his colleagues showedthat certain brain regions are consistently more active during rest than during cognitivetasks
. In 2001, Raichle’s group used brainscans obtained through positron emissiontomography to show a ‘default mode’ of brainfunction related to a physiological baselinewhen the mind is at rest
. That notion gainedfurther momentum two years later when ateam led by Michael Greicius, a neurologistat the Stanford University School of Medicinein California, demonstrated that spontaneousactivity detected by fMRI at rest was strongly correlated across regions within this defaultnetwork
.The default mode network, one of thedozen or so networks in the brain, actually becomes more active during rest than during anovel task. Yet it remains unclear exactly why 
   L   i  z   W   i   l   l   i  a  m  s
“Part of themystique of thedefault networkis no one knowswhat it does.”

You're Reading a Free Preview

/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->