Crucial Advances in 'Brain Reading' Demonstrated

Crucial Advances in 'Brain Reading' Demonstrated

  9:01:00 AM    Science Lover

---

At UCLA's Laboratory of Integrative Neuroimaging Technology, researchers use functional MRI brain scans to observe brain signal changes that take place during mental activity. They then employ computerized machine learning (ML) methods to study these patterns and identify the cognitive state -- or sometimes the thought process -- of human subjects. The technique is called "brain reading" or "brain decoding."

An innovative machine learning method
anticipates neurocognitive changes,
similar to predictive text-entry for
cell phones, Internet search engines.
(Credit: © ktsdesign / Fotolia)

In a new study, the UCLA research team describes several crucial advances in this field, using fMRI and machine learning methods to perform "brain reading" on smokers experiencing nicotine cravings.

The research, presented last week at the Neural Information Processing Systems' Machine Learning and Interpretation in Neuroimaging workshop in Spain, was funded by the National Institute on Drug Abuse, which is interested in using these method to help people control drug cravings.

In this study on addiction and cravings, the team classified data taken from cigarette smokers who were scanned while watching videos meant to induce nicotine cravings. The aim was to understand in detail which regions of the brain and which neural networks are responsible for resisting nicotine addiction specifically, and cravings in general, said Dr. Ariana Anderson, a postdoctoral fellow in the Integrative Neuroimaging Technology lab and the study's lead author.

"We are interested in exploring the relationships between structure and function in the human brain, particularly as related to higher-level cognition, such as mental imagery," Anderson said. "The lab is engaged in the active exploration of modern data-analysis approaches, such as machine learning, with special attention to methods that reveal systems-level neural organization."

For the study, smokers sometimes watched videos meant to induce cravings, sometimes watched "neutral" videos and at sometimes watched no video at all. They were instructed to attempt to fight nicotine cravings when they arose.

The data from fMRI scans taken of the study participants was then analyzed. Traditional machine learning methods were augmented by Markov processes, which use past history to predict future states. By measuring the brain networks active over time during the scans, the resulting machine learning algorithms were able to anticipate changes in subjects' underlying neurocognitive structure, predicting with a high degree of accuracy (90 percent for some of the models tested) what they were watching and, as far as cravings were concerned, how they were reacting to what they viewed.

"We detected whether people were watching and resisting cravings, indulging in them, or watching videos that were unrelated to smoking or cravings," said Anderson, who completed her Ph.D. in statistics at UCLA. "Essentially, we were predicting and detecting what kind of videos people were watching and whether they were resisting their cravings."

In essence, the algorithm was able to complete or "predict" the subjects' mental states and thought processes in much the same way that Internet search engines or texting programs on cell phones anticipate and complete a sentence or request before the user is finished typing. And this machine learning method based on Markov processes demonstrated a large improvement in accuracy over traditional approaches, the researchers said.

Machine learning methods, in general, create a "decision layer" -- essentially a boundary separating the different classes one needs to distinguish. For example, values on one side of the boundary might indicate that a subject believes various test statements and, on the other, that a subject disbelieves these statements. Researchers have found they can detect these believe-disbelieve differences with high accuracy, in effect creating a lie detector. An innovation described in the new study is a means of making these boundaries interpretable by neuroscientists, rather than an often obscure boundary created by more traditional methods, like support vector machine learning.

"In our study, these boundaries are designed to reflect the contributed activity of a variety of brain sub-systems or networks whose functions are identifiable -- for example, a visual network, an emotional-regulation network or a conflict-monitoring network," said study co-author Mark S. Cohen, a professor of neurology, psychiatry and biobehavioral sciences at UCLA's Staglin Center for Cognitive Neuroscience and a researcher at the California NanoSystems Institute at UCLA.

"By projecting our problem of isolating specific networks associated with cravings into the domain of neurology, the technique does more than classify brain states -- it actually helps us to better understand the way the brain resists cravings," added Cohen, who also directs UCLA's Neuroengineering Training Program.

Remarkably, by placing this problem into neurological terms, the decoding process becomes significantly more reliable and accurate, the researchers said. This is especially significant, they said, because it is unusual to use prior outcomes and states in order to inform the machine learning algorithms, and it is particularly challenging in the brain because so much is unknown about how the brain works.

Machine learning typically involves two steps: a "training phase" in which the computer evaluates a set of known outcomes -- say, a bunch of trials in which a subject indicated belief or disbelief -- and a second, "prediction" phase in which the computer builds a boundary based on that knowledge.

In future research, the neuroscientists said, they will be using these machine learning methods in a biofeedback context, showing subjects real-time brain readouts to let them know when they are experiencing cravings and how intense those cravings are, in the hopes of training them to control and suppress those cravings.

But since this clearly changes the process and cognitive state for the subject, the researchers said, they may face special challenges in trying to decode a "moving target" and in separating the "training" phase from the "prediction" phase.

Read More

Brain Cells Responsible for Keeping Us Awake

Brain Cells Responsible for Keeping Us Awake

  11:46:00 PM    Science Lover

---

Bright light arouses us. Bright light makes it easier to stay awake. Very bright light not only arouses us but is known to have antidepressant effects. Conversely, dark rooms can make us sleepy. It's the reason some people use masks to make sure light doesn't wake them while they sleep.

Researchers have identified the group of neurons
that mediates whether light arouses us and keeps
us awake, or not. (Credit: iStockphoto/Osman Safi)

Now researchers at UCLA have identified the group of neurons that mediates whether light arouses us -- or not. Jerome Siegel, a professor of psychiatry at the Semel Institute for Neuroscience and Human Behavior at UCLA, and colleagues report in the current online edition of the Journal of Neuroscience that the cells necessary for a light-induced arousal response are located in the hypothalamus, an area at the base of the brain responsible for, among other things, control of the autonomic nervous system, body temperature, hunger, thirst, fatigue -- and sleep.

These cells release a neurotransmitter called hypocretin, Siegel said. The researchers compared mice with and without hypocretin and found that those who didn't have it were unable to stay awake in the light, while those who had it showed intense activation of these cells in the light but not while they were awake in the dark.

This same UCLA research group earlier determined that the loss of hypocretin was responsible for narcolepsy and the sleepiness associated with Parkinson's disease. But the neurotransmitter's role in normal behavior was, until now, unclear.

"This current finding explains prior work in humans that found that narcoleptics lack the arousing response to light, unlike other equally sleepy individuals, and that both narcoleptics and Parkinson's patients have an increased tendency to be depressed compared to others with chronic illnesses," said Siegel, who is also a member of the UCLA Brain Research Institute and chief of neurobiology research at the Sepulveda Veterans Affairs Medical Center in Mission Hills, Calif.



Prior studies of the behavioral role of hypocretin in rodents had examined the neurotransmitter's function during only light phases (normal sleep time for mice) or dark phases (their normal wake time), but not both. And the studies only examined the rodents when they were performing a single task.

In the current study, researchers examined the behavioral capabilities of mice that had their hypocretin genetically "knocked-out" (KO mice) and compared them with the activities of normal, wild-type mice (WT) that still had their hypocretin neurons. The researchers tested the two groups while they performed a variety of tasks during both light and dark phases.

Surprisingly, they found that the KO mice were only deficient at working for positive rewards during the light phase. During the dark phase, however, these mice learned at the same rate as their WT littermates and were completely unimpaired in working for the same rewards.

Consistent with the data in the KO mice, the activity of hypocretin neurons in their WT littermates was maximized when working for positive rewards during the light phase, but the cells were not activated when performing the same tasks in the dark phase.

"The findings suggest that administering hypocretin and boosting the function of hypocretin cells will increase the light-induced arousal response," Siegel said. "Conversely, blocking their function by administering hypocretin receptor blockers will reduce this response and thereby induce sleep."

Further, Siegel noted, "The administration of hypocretin may also have antidepressant properties, and blocking it may increase tendencies toward depression. So we feel this work has implications for treating sleep disorders as well as depression."

Other authors on the study included Ronald McGregor (first author), Ming-Fung Wu, Grace Barber and Lalini Ramanathan, all of UCLA, the Veterans Affairs Greater Los Angeles Healthcare System and the UCLA Brain Research Institute.

The research was supported by the National Institutes of Health and the Medical Research Service of the Department of Veterans Affairs. The authors report no conflict of interest.

Read More

NASA's Swift and Hubble Probe Asteroid Collision Debris

NASA's Swift and Hubble Probe Asteroid Collision Debris

  2:26:00 AM    Science Lover

---

Late last year, astronomers noticed an asteroid named Scheila had unexpectedly brightened, and it was sporting short-lived plumes. Data from NASA's Swift satellite and Hubble Space Telescope showed these changes likely occurred after Scheila was struck by a much smaller asteroid.

Top: Faint dust plumes bookend asteroid (596) 
Scheila, which is overexposed in this composite. 
Visible and ultraviolet images from Swift's UVOT 
(circled) are merged with a Digital Sky Survey 
image of the same region. The UVOT images 
were acquired on Dec. 15, 2010, when the asteroid 
was about 232 million miles from Earth. Bottom: 
The Hubble Space Telescope imaged (596) Scheila 
on Dec. 27, 2010, when the asteroid was about 
218 million miles away. Scheila is overexposed 
in this image to reveal the faint dust features. 
The asteroid is surrounded by a C-shaped cloud 
of particles and displays a linear dust tail in this 
visible-light picture acquired by Hubble's Wide 
Field Camera 3. Because Hubble tracked the asteroid 
during the exposure, the star images are trailed. 
(Credit: Top: NASA/Swift/DSS/D. Bodewits (UMD) / 
Bottom: NASA/ESA/D. Jewitt (UCLA))

"Collisions between asteroids create rock fragments, from fine dust to huge boulders, that impact planets and their moons," said Dennis Bodewits, an astronomer at the University of Maryland in College Park and lead author of the Swift study. "Yet this is the first time we've been able to catch one just weeks after the smash-up, long before the evidence fades away."

Asteroids are rocky fragments thought to be debris from the formation and evolution of the solar system approximately 4.6 billion years ago. Millions of them orbit the sun between Mars and Jupiter in the main asteroid belt. Scheila is approximately 70 miles across and orbits the sun every five years.

"The Hubble data are most simply explained by the impact, at 11,000 mph, of a previously unknown asteroid about 100 feet in diameter," said Hubble team leader David Jewitt at the University of California in Los Angeles. Hubble did not see any discrete collision fragments, unlike its 2009 observations of P/2010 A2, the first identified asteroid collision.

The studies will appear in the May 20 edition of The Astrophysical Journal Letters and are available online.

Astronomers have known for decades that comets contain icy material that erupts when warmed by the sun. They regarded asteroids as inactive rocks whose destinies, surfaces, shapes and sizes were determined by mutual impacts. However, this simple picture has grown more complex over the past few years.

During certain parts of their orbits, some objects, once categorized as asteroids, clearly develop comet-like features that can last for many months. Others display much shorter outbursts. Icy materials may be occasionally exposed, either by internal geological processes or by an external one, such as an impact.

On Dec. 11, 2010, images from the University of Arizona's Catalina Sky Survey, a project of NASA's Near Earth Object Observations Program, revealed Scheila to be twice as bright as expected and immersed in a faint comet-like glow. Looking through the survey's archived images, astronomers inferred the outburst began between Nov. 11 and Dec. 3.

Three days after the outburst was announced, Swift's Ultraviolet/Optical Telescope (UVOT) captured multiple images and a spectrum of the asteroid. Ultraviolet sunlight breaks up the gas molecules surrounding comets; water, for example, is transformed into hydroxyl and hydrogen. But none of the emissions most commonly identified in comets, such as hydroxyl or cyanogen, show up in the UVOT spectrum. The absence of gas around Scheila led the Swift team to reject scenarios where exposed ice accounted for the activity.

Images show the asteroid was flanked in the north by a bright dust plume and in the south by a fainter one. The dual plumes formed as small dust particles excavated by the impact were pushed away from the asteroid by sunlight. Hubble observed the asteroid's fading dust cloud on Dec. 27, 2010, and Jan. 4, 2011.

The two teams found the observations were best explained by a collision with a small asteroid impacting Scheila's surface at an angle of less than 30 degrees, leaving a crater 1,000 feet across. Laboratory experiments show a more direct strike probably wouldn't have produced two distinct dust plumes. The researchers estimated the crash ejected more than 660,000 tons of dust -- equivalent to nearly twice the mass of the Empire State Building.

"The dust cloud around Scheila could be 10,000 times as massive as the one ejected from comet 9P/Tempel 1 during NASA's UMD-led Deep Impact mission," said co-author Michael Kelley, also at the University of Maryland. "Collisions allow us to peek inside comets and asteroids. Ejecta kicked up by Deep Impact contained lots of ice, and the absence of ice in Scheila's interior shows that it's entirely unlike comets."

NASA's Goddard Space Flight Center in Greenbelt, Md., manages Hubble and Swift. Hubble was built and is operated in partnership with the European Space Agency. Science operations for both missions include contributions from many national and international partners.

For more information, video and images associated with this story, visit: http://svs.gsfc.nasa.gov/goto?10747

Read More

Optical Microscope Without Lenses Produces High-Resolution 3-D Images on a Chip

Optical Microscope Without Lenses Produces High-Resolution 3-D Images on a Chip

  8:43:00 AM    Science Lover

---

UCLA researchers have redefined the concept of a microscope by removing the lens to create a system that is small enough to fit in the palm of a hand but powerful enough to create three-dimensional tomographic images of miniscule samples.

Lens-free tomographic imaging: Schematic diagram 
of the lens-free tomography setup showing the angles of 
rotation for the light source to illuminate a sample. 
(Credit: Image courtesy of UCLA)

The advance, featured in the early online edition of the journal Proceedings of the National Academy of Sciences, represents the first demonstration of lens-free optical tomographic imaging on a chip, a technique capable of producing high-resolution 3-D images of large volumes of microscopic objects.

"This research clearly shows the potential of lens-free computational microscopy," said Aydogan Ozcan, senior author of the research and an associate professor of electrical engineering at UCLA's Henry Samueli School of Engineering and Applied Science. "Wonderful progress has been made in recent years to miniaturize life-sciences tools with microfluidic and lab-on-a-chip technologies, but until now optical microscopy has not kept pace with the miniaturization trend."

An optical imaging system small enough to fit onto an opto-electronic chip provides a variety of benefits. Because of the automation involved in on-chip systems, scientific work could be sped up significantly, which might have a great impact in the fields of cell and developmental biology. In addition, the small size not only has great potential for miniaturizing systems but also leads to cost savings on equipment.

The optical microscope, invented more than 400 years ago, has tended to grow larger and more complex as it has been modified to image ever-smaller objects with better resolution. To address this lack of progress in miniaturization, Ozcan's research group -- with graduate student Serhan Isikman and postdoctoral scholar Waheb Bishara as lead researchers -- developed the new tomographic microscopy platform through the next evolution of a lens-free imaging technology the group created and has been improving for years.

Ozcan, a researcher at the California NanoSystems Institute at UCLA, makes the analogy that a traditional optical microscope is like a huge set of pipes delivering content, in the form of images, to the user. Over years of development, bottlenecks occur that impede further improvement. Even if one part of the system -- that is, one bottleneck -- is improved, other bottlenecks keep that improvement from being fully realized. Not so with the lens-free system, according to Ozcan.

"Lens-free imaging removes the pipes altogether by utilizing an entirely new design," he said.

The system takes advantage of the fact that organic structures, such as cells, are partially transparent. So by shining a light on a sample of cells, the shadows created reveal not only the cells' outlines but details about their sub-cellular structures as well.

"These details can be captured and analyzed if the shadow is directed onto a digital sensor array," Isikman said. "The end result of this process is an image taken without using a lens."

Ozcan envisions this lens-free imaging system as one component in a lab-on-a-chip platform. It could potentially fit beneath a microfluidic chip, a tool for the precise control and manipulation of sub-millimeter biological samples and fluids, and the two tools would operate in tandem, with the microfluidic chip depositing and subsequently removing a sample from the lens-free imager in an automated, or high-throughput, process.

The platform's 3-D images are created by rotating the light source to illuminate the samples from multiple angles. These multiple angles also allow the system to utilize tomography, a powerful imaging technique. Through the use of tomography, the system is able to produce 3-D images without sacrificing resolution.

"The field of view of lens-based microscopes is limited because the lens focuses on a narrow area of a sample," Bishara said. "A lens-free microscope has both a much larger field of view and depth of field because the imaging is done by the digital sensor array and is not constrained by a lens."

The research was funded by grants from the National Science Foundation, the U.S. Office of Naval Research and the National Institutes of Health and was also supported by the Gates Foundation and the Vodafone Americas Foundation.

For more information on the Ozcan research group, visit http://innovate.ee.ucla.edu/.

Read More