The human brain, weighing about 1400 grams, is perhaps the most complex and sophisticated machine in our world. It is extremely diverse in function, carrying almost all our intelligent activities (language, consciousness, thinking and cognition); but it is extremely singular in structure, a neuron connected by synapses Network (contains tens of billions of neurons and trillions of synapses).
Illustration: Brain
To study the brain, we must first have an instrument to observe it. Now the mainstream brain imaging / neuroimaging techniques are: CT, PET, MRI, fMRI and FHIRM-TPM. FHIRM-TPM is a miniature two-photon microscopy imaging system, which was jointly developed and selected as one of the top ten scientific advances in China in 2017. This article will briefly introduce the characteristics of each imaging technology and its specific application scenarios. The depth of the content is limited. Readers who want to know more can read the reference links at the end of the article.
CAT scans, MRI, PET scans, and fMRI. Some neuroimaging techniques allow us to see the structure of the brain, while others allow us to look at brain activity or function.
CT/Computed Tomography scans
the brain by rotating X-rays on a single plane (tomography). Since different brain tissues have different absorption capabilities for X-rays, images of the brain slice can be constructed. By stacking the brain scan images of each layer, we can construct a stereoscopic image of the brain.
CT technology belongs to structural imaging technology, because it can only be used to observe the static structure of the brain, and cannot be used to observe the dynamic function of the brain. The resolution of CT images is not high, but it is sufficient to visualize the main structures of the brain, so it can be used to observe brain tumors.
Note: Ordinary X-ray won the Nobel Prize in Physics in 1901, and CT technology won the Nobel Prize in Medicine in 1979.
图片来源：Efficacy assessment of pemetrexed treatment of an NSCLC case with brain metastasis
Computerized tomography, also known as computerized axial tomography, is a type of structural neuroimaging. It is usually called a CAT or CT scan.
It involves taking a series of x-ray images from various locations around the head. These images can then be combined to construct an image of the brain.
The resolution of CT images is not that high, but they can visualize any major structural problems with the brain, like a tumor.
MRI/Magnetic Resonance Imaging
MRI, like CT, is a structural imaging technology, but MRI no longer uses X-rays, but electromagnetic waves, so MRI is considered to be a safe, fast, high spatial resolution without any harm to the human body method of clinical diagnosis.
The general principle of MRI: when an object is placed in a magnetic field and irradiated with appropriate electromagnetic waves, the rotational arrangement direction of hydrogen atoms (and other atoms, such as oxygen atoms can also be selected) can be changed to resonate, and then we can analyze This process releases electromagnetic waves. Due to the difference in the proportion of water in various tissues in the brain, that is, the number of hydrogen nuclei in the brain, there is a difference in the MRI signal intensity between different tissues. Using this difference as a feature quantity, various tissues can be separated. Similarly, MRI can be used to examine brain structures and visualize tumors.
There is no absolute difference between CT and MRI, in some cases they can complement each other to make up for their shortcomings.
Note: There have been 6 Nobel Prizes (physics, chemistry, physiology or medicine) in the field of nuclear magnetic resonance, which is enough to illustrate the importance of this field and its derived technologies.
Magnetic resonance imaging, or MRI, involves applying a combination of magnetic fields and radiofrequency energy waves to the brain.
Hydrogen atoms respond to the magnetic fields and radiofrequency pulses by emitting energy. The MRI machine receives this energy and can tell what part of the brain it came from.
A computer can use that information to reconstruct an image of the brain that has high spatial resolution.
Image source: wiki
PET/Positron Emission Computed Tomography
The most obvious feature of PET technology is that the test object needs to take a contrast agent (usually fluorinated deoxyglucose, fluorine-18) labeled with a radiotracer isotope with a short half-life (basically non-toxic), and after a period of time, the contrast agent into the metabolic circulation of the whole body.
The characteristic of radioisotopes is that positron emission decay occurs, releasing a positron (ie, an electron corresponding to an antiparticle), and the positron encounters an electron in the organism and produces an electron pair annihilation. This signal can be used. PET scanner capture. Because the contrast agent can persist throughout the brain, we can obtain three-dimensional and functional images of the entire brain.
Unlike CT and MRI, which can look directly at the structure of the brain, PET usually looks at brain function indirectly by looking at blood flow, oxygen consumption, and tracking neurotransmitters. When an area of the brain is active, blood flow and oxygen consumption in that area is accelerated, and since the contrast agent is continuously present in the blood circulation of the brain, the dynamic changes in the brain can be observed using PET.
In addition, because malignant tumors metabolize glucose much faster than benign tumors, PET can be used clinically to distinguish benign tumors from malignant tumors, as shown in the figure below.
Positron emission tomography, or PET scanning, is a way of imaging brain function. To do a PET scan, a patient is injected with a radioactive substance that emits positrons,
which then emit gamma rays when they collide with electrons in brain tissue. These gamma rays are detected by the PET scanner. Because the radioactive substance was injected into the bloodstream,
what the PET scanner is detecting is the movement of blood throughout the brain. Blood flow to an area of the brain increases when that area is active, so PET scanning creates an image that highlights the areas of the brain that are being used the most while the person is in the scanner.
Image source: Positron emission tomography
fMRI/functional magnetic resonance imaging
fMRI absorbs the technical advantages of MRI and PET, and realizes brain functional imaging by detecting changes in the magnetic field of blood flow into brain cells, thus extending the original structural imaging technology MRI to functional imaging.
When a neuron is active, the blood flow in its vicinity is accelerated to replenish the consumed oxygen, so nerve activation can cause changes in hemodynamics. BOLD (Blood oxygen-level dependent) is a commonly used measurement index in fMRI. It describes the ratio of oxygenated/deoxygenated heme in the blood. When neurons are activated, the ratio of oxygenated heme increases, and the relative BOLD signal also increases. Then strengthen. Heme is diamagnetic in its oxidized state (oxygenated heme) and paramagnetic relative to deoxygenated heme, so changes in neuronal activity can be captured by high spatial resolution MRI.
Since fMRI can continuously detect activity signals in the cerebral cortex, it has been widely used in research fields such as brain function localization and cognitive psychology.
Functional MRI, or fMRI, uses a similar approach to MRI but focuses on the different responses oxygenated and unoxygenated blood make to magnetic fields and radiofrequency energy.
fMRI uses what is called blood-oxygen-level-dependent contrast, or BOLD, to identify changes in blood flow in the brain, and thus to identify areas of the brain that are most active.
fMRI allows one to image brain function without having to inject anything, and it provides high resolution MRI images at the same time as it provides a functional image.
FHIRM-TPM/miniature two-photon microscopy imaging system
"Good progress has also been made in understanding the neural circuit mechanisms that process sensory signals such as vision, hearing, and smell, but we still know very little about complex functions such as learning, memory, attention, and decision-making, let alone common emotion, self-awareness, thinking and language. Language is a unique ability of human beings, and I think the understanding of the neural circuit mechanism behind language processing is one of the most important goals of neuroscience. Language impairment caused by brain injury is a key to the study of language It provides valuable clues about the neural mechanisms of language, but since we can only study the human brain using non-invasive experimental means, it is very difficult to study language in depth."
The spatial resolution of macroscopic human brain imaging is very low, on the order of millimeters. It can be used to determine gross structural and functional changes in the brain, but not to understand the structure and function of neural circuits. Brain imaging methods such as CT, MRI, and PET play an important role in clinical diagnosis, and a better understanding of the connection between MRI signals and the structure and function of neural circuits can help us make better use of MRI technology.
In order to understand what I'm thinking, you need to observe and understand the dynamic patterns of activity of at least millions of neurons in my brain. I'm not sure we can do that in the foreseeable future.
Functional MRI and PET imaging show that many areas of your brain are active, including much of the cerebral cortex. Neuroscience's work in understanding consciousness will be largely complete when we can clearly understand which neural circuits are involved in the generation of consciousness-related brain states, and how they are activated and regulated.
There is a gap between our macroscopic and microscopic understanding: we still know very little about how the complex neural circuits formed by the large number of neurons in the brain process neural information. An important task for future neuroscience is to understand the structure of neural circuits and their regularity of activity when performing various brain functions.
In the next decade, there were two major challenges: one is to observe the activity of large groups of neurons in the living brain simultaneously at the resolution of individual neurons and nerve fibers; Analyzing and interpreting extremely large amounts of data on the dynamic activity of neurons during the process.
Imagine a day when a device could observe the activity of every neuron and synapse in our brains.
Why it is possible to live, more focus on the update of technology, and the highlights of this technology
The research group of Cheng Heping and Chen Liangyi of the State Key Laboratory of Biofilm and Membrane Bioengineering of Peking University has cooperated with Zhang Yunfeng and Wang Aimin of the School of Electronic Engineering and Computer Science to develop a microscopic imaging system that can realize free-state brain imaging. Optics, ultrafast fiber lasers, and semiconductor optoelectronics technologies have made breakthrough technological innovations in the development of high spatiotemporal resolution in-vivo imaging systems, and successfully developed a 2.2-gram miniaturized wearable two-photon fluorescence microscope, which is the first record in the world. High-speed high-resolution images of neuronal and synaptic activity in the mouse brain under natural behavioral conditions such as tail, platform, and social interaction.
This breakthrough technology will open up a new research paradigm and realize long-term observation of multi-scale and multi-level dynamic information processing such as synapses, neurons, neural networks, and multi-brain areas under the natural behavior of animals. It can not only "see" the process of brain learning, memory, decision-making, and thinking, but also play an important role in visualizing the neural mechanisms of autism, Alzheimer's disease, epilepsy and other brain diseases.
The imaging system was called by Edvard I. Moser, winner of the 2014 Nobel Prize in Physiology or Medicine, as a revolutionary new tool for studying the brain's spatially localized nervous system.
Structural models of neurons and protrusions
Top 10 annual progress in Chinese science released
Micro two-photon microscopy imaging system
With the rapid development of imaging technology, it is conceivable that one day, we will be able to capture the activity signal of every neuron in the neural network of the brain in real time. If we can decode it through mathematical modeling or other methods, maybe we can understand human intelligence. generate new awareness.
Once we truly understand intelligence, the arrival of strong artificial intelligence is not far away.
The above is a brief introduction to the current common brain imaging techniques, and there will be opportunities to introduce the application of brain imaging in cognitive psychology and other aspects.
Reference link:
What Is FMRI? - Center for Functional MRI - UC San Diego
Difference Between MRI and fMRI
Overview of Functional Magnetic Resonance Imaging
How Does It Work?: Positron emission tomography - NCBI - NIH
Fast high-resolution miniature two-photon microscopy for brain imaging in freely behaving mice - YouTube
Fast high-resolution miniature two-photon microscopy for brain imaging in freely behaving mice - Nature Methods
A new generation of miniature two-photon fluorescence microscopes
How are memories stored and retrieved in the human brain?
Types of Brain Imaging Techniques
What is so special about the human brain? | Suzana Herculano-Houzel
Human brain mapping and brain decoding. | Jack Gallant | TEDxSanFrancisco
This Giant Neuron Could Explain Where Consciousness Comes From