Friday, December 11, 2020
New Research: NeuroTracker Predicts the Profitability of Traders
With breakthroughs coming faster than in any other field of science, a lot has been going on in neuroscience in recent years. Here are 22 genuinely mind-blowing neuroscience studies that challenge our preconceptions of who we are, or who we could be.
Earlier this year MIT scientists developed a new technique to pair structural mapping (brain anatomy) with functional mapping (how the brain behaves) - the first time this has been properly achieved. In addition, this has been done in live mice, with the mapping performed across mouse brain regions in real-time. This video gives an idea of how fascinating it is to see the coupling of brain structures and live activity changing in response to a mouse being shown different images.
The vanguard technique combines third-harmonic generation (THG) three-photon microscopy with retinotopic mapping, allowing activity to be observed through deep brain tissue via electrical signatures.
It also delivers stunning resolution, allowing individual neurons and their substructures to be studied, as well as fine blood vessels and myelin – a kind of insulator known to be a critical factor in brain processing speed.
This study focused on the visual centers of the brain, but the same method can be used to study other regions. It promises to be a powerful tool for understanding differences in healthy and diseased brain states, as well as how the brain responds to environmental stimulation.
Stanford University made a key breakthrough with a new bifocal microscopy technique called COSMOS. Their work captured movies of neural activity across the whole of the cerebral cortex of a mouse brain.
These signals were recorded by essentially filming the brain from three different angles, then computationally extracting signals to provide a live video of macroscopic activity over the left and right hemispheres. Here is a sample where you literally see the remarkable electrical storm of a real brain in action.
As the cortex handles complex higher-level cognitive functions, more mysterious behaviors like decision-making processes can now start to be unraveled in a global way. For example for understanding of the relationship of decisions dependent on sensory perception and motor function (think about what’s involved in deciding which way to dodge an oncoming car).
The researchers also expect COSMOS to be a low-cost method for screening the effects of psychiatric drugs, so that they can be developed to be more functionally effective.
As we’ve covered in a previous blog, a major breakthrough for Google’s Deep Mind artificial intelligence program came through mimicking the neo-cortical columns of the human mind. This led to vastly increased intelligence using a fraction of the computing power. As a result this human-modelled AI has now surpassed the world’s best chess, Go and then eSports players at their own games.
Though not fully understood, sleep provides a critical function for mammalian and human brains, with serious problems occurring whenever sleep deprivation is endured. This year Los Alamos National Laboratory discovered that the spiking computational networks of AI systems also suffer a kind of sleep deprivation, becoming unstable when performing for long periods without rests. Yet, when put into a network state similar to the brainwaves we experience during sleep, optimal performance was restored.
This may not sound like such a big deal, but advancements in AI are likely to transform the way we all our lives. The findings also hint that the merging disciplines of neuroscience and AI field could yield a new era of super smart computers.
A minuscule brain device has been used to improve quality of life patients with severe upper limb paralysis caused by motor neuron disease. Carried out at the University of Melbourne, this trial implanted the new micro technology inside the brains of the participants.
The device called Stentrode™ was inserted through keyhole surgery into the neck, and from there moved into the motor cortex via blood vessels. This minimally invasive method avoids the associated risks and recovery complications of open brain surgery.
The implant uses wireless technology to relay specific neuronal activity into a computer, where it is converted in actions based off the intentions of the patients. Amazingly, this tiny chip allowed the patients to perform actions like click and zoom, and write with 93% accuracy, helping them do things we take for granted like text, email and shop online.
It's very early days still, but the minimally invasive nature of the treatment shows the great potential for micro neurotechnologies to help aid people with all kinds of cognitive impairments.
In 2018 we reported that scientists learned how to reprogram stem cells into specific neurons. This year researchers from four different US universities have taken a bigger step towards the holy grail of life extension. By identifying genes networks that regulate cellular regeneration, they have been able to manipulate normal cells to turn into progenitor cells, which can morph into any cell type to replace dying cells.
Their proof of concept was carried out with the glial cells of Zebra fish, effectively converting them into stem cells which then detected and restored damaged retinal cells to recover impaired vision.
Cell death, or apoptosis, is a plays a big role in the inevitably of natural aging in humans. The researchers believe that the process for regenerating neurons in the brain will be similar. If successful it will have vast implications for conditions such as Alzheimer’s Disease, where large regions of the brain can be lost to the death of neurons. It may also play a role in preventing the many side-effects of natural aging in the brain, for longer and healthier living in peak shape into old age.
Rather than replaced dying cells, scientists at Heidelberg University have identified key processes involved in the death of brain cells, called neurodegeneration. It involved uncovering the process by which cellular glutamate uptake prevents cell death in healthy people, yet becomes inactive in diseased state like stroke, where oxygen supply to brain cells becomes restricted.
In effect this leads to cell killing themselves off simply because they are not getting the correct chemical signals to tell them to stay alive. The researchers then developed a special class of inhibitors that can step in and deactivate the cellular ‘death complex’ before it occurs.
The inhibitors showed to be highly effective at protecting nerve cells, hopefully leading to a new class of treatments options for neurodegenerative diseases.
Aarhus University researchers have used advanced PET and MRI imaging techniques to reveal Parkinson’s disease to actually be either of two different variants of the disease.
In one variant the disease starts in the intestines, going on to spread to the brain through neural connections. In the other, it starts in the brain and then moves into the intestines and other organs. This video gives a great overview.
Though not curative, it’s a major step in the right direction for being able to identify early stage onset for preventative measures. For example, it may lead to treatments which prevent the disease from even making it into the brain altogether, where the effects then become debilitating over time. It is also another key piece in the puzzle of the powerful symbioses between our intestines and our mind, known scientifically as the gut-brain axis.
Scientists at the University of Cambridge and Imperial College London have developed a new type of AI algorithm that can detect, differentiate and identify different types of brain injuries from topographical CT scan data.
CT scans collect a huge amount of data which can take experts hours to analyze, and this needs to include the collective evaluation of multiple scans over time in order to track recovery trajectories or disease progression. This new AI tool appears to better than human experts at detecting such changes, as well as being far quicker and cheaper.
For example, their research showed the software to be highly effective at automatically quantifying the progression of multiple types of brain lesions, helping predict which lesions would get larger. The innovative application of this type of AI to assist human analysis is likely to be first of many that will transform medical diagnostics in cost-effective ways.
Super-agers are individuals whose cognitive skills are way past their peers in old age, retaining youthful mental abilities well into their 70s and 80s. Until now the secret to retaining their peak shape has been little understood.
University Hospital Cologne and the Research Center Juelich have discovered a key difference in their biology. Using PET scans they revealed that super-agers have markedly increased resistance to tau and amyloid proteins. Until recent years these proteins have proven difficult to study.
Super-agers also have lower levels of tau and amyloid pathology, which in turns leads to various kinds of neurodegeneration in most people in their later years. It’s now been identified that reduced resistance to tau and amyloid accumulation is a primary biological factor for the loss of peak cognitive shape.
New research can be focused on these processes to find ways to possibly cure mental decline generally, as well as help develop therapeutics to protect against forms of dementia that are already occurring.
A research team at the University of California San Francisco have successfully developed a method using deep brain stimulation (DBS) to adaptively treat depressive symptoms only when they appear. Deep brain stimulation involves implanting electrodes within the brain to deliver electrical currents to alter brain activity.
Previous studies have had limited success for treating depression with DBS because devices could only deliver constant electrical stimulation in one area of the brain. However depression can affect various areas of the brain, and the neural signatures of depression can rise and fall unpredictably.
With the aim of essentially creating a pacemaker for the brain, the scientists decoded a new neural biomarker. This specific pattern of brain activity effectively predicts the onset of symptoms. With this knowledge the team customized a new DBS technology that only activates when and where it recognizes that pattern.
The type of automatic on-demand therapy is impressive because it's functional responses are unique to both the patient’s brain and the neural circuit causing the illness. In it’s first trial, this custom DBS method was tested with a patient suffering from severe depression and passed with flying colors. Almost immediately, the patient’s symptoms were alleviated, and this continued to be the case long term.
In the COVID era, where anxiety and mental health problems are becoming rife, this approach could prove an invaluable drug-free therapy for hundreds of millions of people.
Similar to light waves, humans can only perceive a relatively small spectrum of the sound waves that travel around us. Typically we can only pick up on frequencies between 20 Hz and 20,000 Hz, beyond this is considered ultrasonic. This is the frequency range that animals like bats operate in, and also what is put to use in ultra sound medical scans.
A new method utilizing sophisticated technology has been pioneered by scientists at Aalto University, and has led to a device that basically gives humans bat-level hearing. This includes not only the ability to hear frequencies well beyond 20,000 Hz, but also to discern the direction and distance of the sound sources. For biologists for example, it allows people to track otherwise stealthy bats in flight, and locate their positions.
It works by recording ultrasound via a spherical microphone array, which detects ultrasonic sounds and uses a computer to translate the pitch to audible frequencies. It then plays the converted sound waves through headphones in real-time. Being able to perceive normally inaudible sounds could have valuable industrial applications, for example being able to hear and locate otherwise silent gas leaks.
Although neuroscience is a relatively young and fast-growing domain of science, artificial intelligence (AI) is both much newer and growing faster. The potential of combining these two fields of science has been revealed by researchers at MIT.
Using machine learning, they have discovered that artificial neural networks can self-learn how to smell in just a few minutes, actually mimicking the olfactory circuits in mammalian brains. This is profound because the algorithm put to work had no knowledge of the millions of years evolution required to develop smell biologically.
Yet amazingly, the artificial neural network replicated the biological activity of smell so closely that it revealed the brain’s olfactory network is mathematically optimized for its function.
This precise mimicking of the natural structure of circuits in the brain by independent machine learning may herald a new era, whereby AI teaches us the inner secrets of biological evolution. Sense of smell is the starting point in 2021, but who knows where this could lead…
Researchers at UC San Francisco developed a new kind of a speech neuroprosthesis for patients with paralyses that prevents them for speaking. The method was demonstrated successfully on a man with a severely damaged brain stem, causing whole body paralysis.
Somewhat remarkably it works by detecting speech-related brain signals that control the vocal cords. When we speak, the vocal cords require complex motor-function instructions in order to articulate the wide variety of sounds we use when conversing. Even when unable to move, these signals can still get sent from the brain.
Using brain recordings from epilepsy patients, the scientists developed a method for real-time decoding of instructions to vocal muscles, into words. From these neural patterns, they were able to reliably discern 50 different common words whenever the patient was thinking them.
All that was required was for the patient to wear a high-density electrode array to capture and record neural activity, which recorded signals from the speech motor cortex. This allowed up 18 words per minute to be translated with 93% accuracy. The advantage for the patient was that he simply had to act like he was really speaking and he could communicate hundreds of different sentences from the 50 words vocabulary.
Although this breakthrough seems limited to paralyzed patients, we undergo paralysis every night when we dream (unless we sleep walk). If evolved sufficiently, this approach could, for example, pave the way to translating our very thoughts while sleeping!
Technically termed ‘brain organoids’, mini-brains can be grown from induced pluripotent stem cells. These stem cells can be taken from a person’s skin or blood, and the the potential to be morph into any type of cells. The benefit is that cell structures normally very difficult to access, can in principle, be grown and isolated for study. This is especially relevant for the brain, however previous mini-brains had limited functional structures.
This year’s breakthrough by scientists at UCLA has catapulted the structural complexity by growing aggregates of organoids to form complex three-dimensional brain structures. The researchers took stem cells from patients with Rett syndrome (a condition with seizures), and were able to grow mini-brains with functional activity similar to parts of human brains. This meant they were able to safely and successfully observe patterns of electrical activity that resemble the onset of seizures.
This research shows for the first time that some aspects of brain function can be isolated and studied in the lab down to the level of individual living cells. The key advantage is that these mini-brains can be grown to replicate aspects of both normal and diseased brain functions, as well as to test drugs and treatments with no risks to human or animals.
The scale of the human brain is enormous, so there are still clear limitations in terms of the complexity of brain structures that can be studied, but clearly this emerging neuroscience domain has sci-fi like potential.
With the exponential growth in growth in computing power in recent decades, microchips have been getting increasingly smaller each year. Tech focused neuroscientists at Brown University have now developed a wireless computer so small it can be easily missed by the human eye. Dubbed ‘neurograins’ - because they are about the size of a grain of salt - they were developed to track and monitor brain activity.
These ultra-tiny computers are able to record electrical activity from nearby neurons, and transmit their data wirelessly. The goal was to develop a new type of brain-computer interface (BCI) system, where a network of the mini-sensors can collectively track meaningful aspects of brain activity, and send the information to a nearby hub.
In a proof-of-concept experiment, the researchers deployed a network to successfully record a rodent’s neural activity with much greater accuracy than ever achieved before. This recording of brain signals in unprecedented detail it’s still in it’s early stages, but the technological breakthrough holds much promise for being able to convert brain waves into useful real-world actions without any physical effort.
This year a new type of microelectrode array has been used to create a form of artificial vision via a visual prosthesis. University of Utah scientists at the John A. Moran Eye Center built the device to record and stimulate neuronal activity within the visual cortex.
Implanted within the eye, the array receives visual information through glasses containing a small video camera, with the data processed by specialized software. The device then activates retinal neurons to produce phosphenes, as if they are receiving points of light. In turn allowing basic images of lines and shapes to be percieved by the mind.
Trialed with a completely blind patient, this method proved to be effective, and involved no complications from the surgery or the neuronal stimulation. In this first test, only a single array was used. However, the next goal is to use 7 to 10 arrays to deliver more detailed images that will allow blind people to actually navigate the world visually.
A new class of ‘dancing molecules’ has been applied by researchers at Northwestern University to repair tissue in severe spinal cord injuries and successfully reverse paralysis. The dancing part involves manipulating the motion of these molecules to that they can wiggle their way into normally impossible to reach cellular receptors, in order to prompt them to get into gear repairing nerve tissues.
These seemingly magic molecules work by setting-off cascading signals, triggering axons to regenerate and helping neurons to survive after injury by encouraging a variety of new cell types to be born. This is in turn supports the regrowth of lost blood vessels necessary for cellular healing.
Tested in mice, just a single injection of the molecular therapy led to the paralyzed mice being able to walk again in under four weeks. Somewhat conveniently, 12 weeks later (well after recovery is complete), the materials biodegrade into nutrients for the cells without any side effects, effectively disappearing from the body naturally.
Virtual Reality (VR) has been used by psychophysicists for decades to investigate how we perceive sensory information. This year researchers from the University of Basel, the oldest university in Switzerland, developed a virtual reality application to actually treat height phobias.
Called Easyheights, the smartphone compatible software provides exposure therapy using 360° images of real locations. Wearing a VR headset, users stand on a platform that starts one meter above the ground, and then progressively rises as the users acclimatizes to each stage of height. It works by increasing sensory exposure to height with without increasingly the level of fear.
A clinical trial demonstrated the efficacy of this immersive form of treatment, producing significant reductions in phobia in real height situations. The benefits were experienced with just four hours of home-based training. This discovery shows how combining neuroscience knowledge with today’s technologies, can clinically improve peoples’ quality of life in ways that are easily accessible.
As we speak, neuroscientists at the Max Planck Institute for Evolutionary Anthropology are literally building “miniature brains” genetically grafted with multiple versions of Neanderthal DNA. Using the bottom-up futuristic biotech known as CRISPR, these lentil-sized mini-brains will contain clusters of live neurons grown from stem cells, performing real brain activity.
Although they will be too small to involve any complex behavior like communication, it is expected that they will reveal differences in fundamental brain activity that Neanderthals may have had. In this way genetics is providing a kind of historical telescope for neuroscience, allowing it to peer into the workings of ancient brains. All this from DNA preserved in bone fragments for tens of thousands of years.
And if you think this is something as simple as a few cells in a petri dish…think again. The German researchers are planning to hook-up the Neanderthal mini-brains to robots, in order to observe behavioral outputs. Even more ambitious than the plot a futurist sci-fi movie, if successful the mind simply boggles at what will be possible in the coming years – Neanderthal robot house maids anyone?!
One of the biggest challenges neuroscientists face is that it is very difficult to study live brains. Even with brains recently deceased, neurons rapidly decompose in the hours after death, literally disintegrating. To tackle this challenge gung-ho neuroscientists at Yale University created a vanguard biotech called BrainEx. This high-tech support system was designed to keep brain cells alive in the way that hair and finger nails keep growing post-mortem.
Putting the tech to the test, the researchers used BrainEx to restore synaptic activity and circulation to a pig brain that had been dead for four hours. The brain had been removed from the pig and revived with an artificial blood supply using a proprietary mixture of protective, stabilizing and contrast agents. This took place just before the destruction of cellular and molecular functions started to take place. The image below shows the difference between a normally disintegrating pig brain 10 hours after death (left), and health looking cells on the revived pig brain (right).
Here comes the zombie part. Although the neurons were being kept alive and kicking, there was no higher-level functional activity in the brain circuits – so alive and dead at the same time. This flip from Frankenstein-like fiction to non-fiction, shows how neuroscience can change big ethical questions from the philosophical to the practical.
The biotech isn’t limited to zombie pigs though, in principle it will work with any kind of mammalian brains…including humans! The breakthrough has huge potential for improving our working knowledge of how our own minds operate. At the same time, it does looks unnervingly close to bringing the dead back to life.
On a more inspiring note, 2019 also saw the development of a computer system capable of translating brain activity into synthesized speech. It works by decoding the movements of muscles involved in speech via nerve impulses analyzed through electrophysiological activity. The results of an experiment at the at the University of California San Francisco showed that a prototype version could successfully interpret language through muscular nerve signals, if speaking slowly.
The researchers expect to improve the biotech to natural speech speeds, which are around 150 words per minute. Still, it is already quite remarkable considering that only brain signals are measured. Here is a video demonstrating how patterns of brain activity from the speaker’s somatosensory cortex, decoded into vocal tract movements, can then be interpreted as language.
Many scientists have tried to solve this problem before and failed. These researchers took a fresh approach by creating artificial intelligence models for building simulations of vocal tracts. In effect the AI then taught itself from a library of speech experiments data and trained its neural networks to be able to decode language from vocal movements. These developments could be important steps in simulating human biology in computer programs for research purposes.
From a medical perspective, many patients with throat or neurological conditions, such as strokes or paralysis, can completely lose their abilities for speech. This neurotechnology paired with a smartphone could allow the voiceless to talk normally in real-time, on an everyday basis, simply by thinking about speaking.
However, as the simulated voice only requires reading a small region of brain activity, and the speech could be sent to virtually any computer, then potentially anyone could silently and covertly communicate to anyone with a smartphone and headphones. As that system could be two-way, it represents a literal neurotech solution for human telepathy. The possibilities are endless.
NeuroTrackerXNeuroTrackerX is a cognitive training program designed to assess and improve mental performance.
*Elite athletes and skilled specialists from teams and organizations like these. All trademarks and logos are intellectual property and owned by the respective organizations listed, not NeuroTracker.*
** NeuroTracker is used in various peer reviewed medical research and applications under the guidance of a licensed medical professional. NeuroTracker is not intended to be substituted for professional medical advice, diagnosis, or treatment.
