Please enable JavaScript.
Coggle requires JavaScript to display documents.
Biopsych - RM; Brain Imaging (Y2) - Coggle Diagram
Biopsych - RM; Brain Imaging (Y2)
History of imaging techniques
Structural techniques -
Radiographic (X-rays) - view the inside of objects
Impact on society is immense and applied across fields
-> Medical, food inspection, security and archeology
Considerations -
Invasiveness
Suitable to measure structure or function
Indirect or direct information about the brain
Radiography
Use of X-Rays to view a cross sectional area of a non-uniform material such as the human body
Beam of X-rays produced by an X ray generator is projected towards an object
Based on the density of the different areas of the object a proportion of X-rays is absorbed - bones are too dense for the rays to pass through
X-rays that pass through are captured behind the object by a detector
Gives 2D representations of the structure
Arterial imaging -
Walter Dandy - performed first clipping of an intracranial aneurysm, marking the birth of cerebrovascular neurosurgery
Visualisation of blood vessels in the brain
Catheter inserted into a large artery e.g. femoral and threaded to carotid
Contrast agent injected
X-rays taken
Ventriculography -
Dandy (1919) imaged brain’s ventricular system
Filled with CSF
Filtered air directly into the lateral ventricles of the brain via small holes drilled into the skull under anaesthesia
Not painful, but not safe (risk of haemorrhage and infection)
Pneumoencephalography -
Drain CSF and replace with air, oxygen or helium
Causes change in density
Enhanced X-Ray
Extended scope for intracranial diagnosis - high risk, painful, long recovery period and invasive
Modern imaging diagnostic techniques
Contrast imaging remains essential part of neurosurgeons clinical diagnostic imaging and used in the management of aneurysms and some types of brain tumours
Techniques have become refined in past few decades
In the past few decades, we have seen advancement in the different types of neuroimaging methods utilised in both medical diagnoses and research
Method depends on question -
Structure - CT and MRI
Temporal processing - Single or multi-unit recording, EEG and ERP
Function - TMS, PET, fMRI
Invasive / noninvasive -
Invasive - single unit recording and PET (positron emission tomography)
Non-invasive (radiation) - X-ray and CT
Non-invasive (no radiation) - Multi-unit recording (EEG), TMS, MRI and fMRI
CT scan - computerised tomography
Developed by Oldendorf (1961) and further by Hounsfield and Cormack (1973) who won the Nobel Prize in Physiological and Medicine in 1979
Measures the absorption of an X-Ray photon beam through the patient, similar to X-ray imaging
X-ray beam rotated around the patient, gives information about the photon absorption from different angles
By combining measured absorption profiles from various angles, it is possible to create a cross-sectional image
Different tissue (bone, blood, soft tissue, lungs) have different absorption rates known as attenuation coefficients, providing a cross sectional image
Used now for both diagnostic and research
Safe, painless, non-invasive and repeatable
MRI - Magnetic Resonance Imagery -
Random direction of magnetic dipole moments in absence of external magnetic field
Alignment of magnetic dipole in presence of external magnetic field (parallel and antiparallel)
Use of radio frequency pulse to flip the magnetic dipole moments into the transverse plane - generated with transmitter coil
Precession of magnetic dipoles within the transverse plane, measurement of the radiofrequency signal caused by the precession using a receiver coil
Three planes of brain images - transverse, coronal and sagittal
Three anatomical planes - sagittal, coronal and transverse
TMS - Transcranial Magnetic Stimulation -
Noninvasive method used to stimulate small regions of the brain
Magnetic field generator - coil - produces electric currents
A pulse generator delivers electric current to the coil
Used clinically to evaluate connection between brain and muscle
Approved for migraine and major depressive disorder treatment
A pulse generator delivers electric current to the coil
Used in research to map cortical region function and cause virtual lesions and impairment
Early functional neuroimaging -
Initially used for reflecting brain activation from speaking, reading, visual or auditory perception and voluntary movement
Early techniques used radioactivity
Xenon inhalation was used for first blood flow maps of the brain
Developed by Lassen, INgvar and Skinhoj (1960s)
Functional imaging revolution -
-> Began with PET - positron emission tomography is a molecular imaging technique that measures the distribution of a radioactive tracer
Radio ligands -
-> Radioactive substance which enter the bloodstream to the brain and bind to receptors
-> Mainly water molecules because oxygen is taken up by active brain regions
-> Often O15 used to trace the receptor / uptake system in the body
-> Radio-ligands are either single photon or positron emitters
-> When the radioactive isotope in the ligand decays it can be measured by positron emission tomography
-> PET analyses - Subtraction method - progress use of oxygen-15 labelled water; Active neurons recruit H2015 allowing regional maps of brain activity to be made during cognitive tasks
Functional MRI
Scientists learned that the large blood flow changes measured by H20-15 PET could also be imaged by MRI
Functional Magnetic Resonance Imaging (fMRI) was born
Since the 1990s, fMRI dominates the brain mapping field due to its many advantages
-> Low invasiveness
-> Lack of radiation exposure
-> High spatial resolution
-> Wide availability
The BOLD signal -
When neurons are active, they increase consumption of energy from glucose to glycolysis (more rapid uptake)
The response is an increase in blood flow to the region of increased neural activity
Blood-oxygen level dependent (BOLD) is the MRI contrast of blood deoxyhemoglobin (1991)
MRI can detect these changes because haemoglobin has magnetic properties that change slightly depending on the levels of oxygen
Activations -
Statistical parametric map - shows areas of significant difference / correlations
Calculated with correlations of behaviour and comparing groups on their grey and white matter counts
Data analysis issue -
Susceptibility to artefacts
Spatial normalisation
Spatial smoothing
Partial volumizing
Normalising the image -
Spatial normalisation - fitting an image from native space into MNI space, or a template
Smoothing the image -
Primary reason - increase signal to noise ratio
Effect - data becomes more normally distributed
Compensates for inexact nature of spatial normalisation - smoothes out incorrect registration
Diagnostic and research tool -
MRI and CT scanning play large role in medicine because they are still largely focused on structure
fMRI and PET are still largely devoted to research because they focus on function
Recently started to use fMRI to being to answer clinical questions
Likewise, PET is increasingly used for clinical diagnoses
EEGs
Berger (1929) reported the first successful recording of electrical sleep activity from the human scalp
Neurophysiologists quickly confirmed that Berger’s recordings actually represented activity generated by the brain
Berger’s electrically recorded neural activity became known as the electroencephalogram (EEG)
How do we know that EEG is neural activity?
Within the cellular makeup of the brain, cells possess fluid both inside and immediately surrounding them separated by a membrane
Changes in these fluids can cause changes in the electrical state of the cell
They are responsible for the transmission of information within cells and between cells
What is measured with EEG:
Ionic current flows generate two types of electrical potentials for information transmitted within a cell or between cells
Action potentials are transmission of information within a single cell, from the cell body to the axon terminals via the cell’s axon
Postsynaptic potentials are the transmission of information between neurons from a cell’s membrane via the apical dendrites (spiny appendages protruding from the cell body)
Neurons (action potentials) -
When an action potential arrives at the axon terminal of the presynaptic axon - it causes the release of neurotransmitter molecules that open ion channels in the postsynaptic neurons dendrite
The combined excitatory and inhibitory postsynaptic potentials of such inputs can begin a new action potential in the postsynaptic potentials
EEG reflects postsynaptic potentials from the dendrites of cortical neurons (Allison et al, 1986)
PSPs can be inhibitory (IPSPs) or excitatory (EPSPs), but EEG mainly represents EPSPs
Originate from cortical pyramidal cells because they generally obey a parallel orientation
Recording EEG - Passive single and multi unit processing
Invasive - implanting an electrode
-> Usually only done with animals or people requiring neurosurgery
Passive - recording the activity (action potentials) of one or more neurons, not stimulating activity
Evaluation of single cell recording -
Process concerned with some stimuli e.g. lights for the visual cortex
Highest spatial resolution
Can follow activity over time
Low generalisability
High invasiveness
Only used on animals
Sheds light on basic functions of individual cells - how mirror neurons were located
Pyramidal alignment - perpendicular to scalp -
These neurons are good for scalp measuring because
-> Oriented with their dendrite at the cortical surface, and axons forming the white matter deep in the cortex
-> The proximity of cortical regions to the scalp influences the strength of the signals
-> However, requires contribution of many 1000s of neurons to get signal to get signal big enough to see
-> Amplitude of signal therefore depends on synchronicity of activity in underlying neurons -> rhythms / waves of synchronous excitation
Electro caps - non invasive placement of electrodes on scalp, which are amplified and filtered
Standard system generated placement (Jasper, 1958) - midline, frontal, parietal, temporal and occipital regions
-> Purpose - create a way to place electrodes consistently over the same cortical areas across subjects and patients, regardless of variations in head size
Electrodes connect to fixed points on scalp using conductive gel - these in turn are connected to banks of amplifiers and recording devices
Reference electrode used on a site such as the bone - EEG calculates potential difference between two electrode locations
A single reference electrode connects to each of the recording electrodes on the scalp
Source localisation -
Electrodes record an average of neural activity of neighbouring areas across scalp locations
But, electrical activity from one area can be detected at distant and multiple locations
Cannot assume that EEG scalp activity is generated from the brain areas directly under the electrode site
Human and animal studies help make predictions about locations of EEG generators
Mathematical algorithms have been developed in attempts to localise the origins
Temporal resolution and rhythms -
We get extremely rich temporal information from EEG
Scalp EEG activity shows oscillations at a variety of frequencies
Oscillations have characteristic frequency ranges associated with different states of brain functioning
High frequency / low amplitude - alertness + dream stages - beta wave
High amplitude / low frequency - sleep stages (Delta)
EEG clinical applications -
EEG can be used as a diagnostic tool
Recording of the brain’s spontaneous electrical activity over a short period of time (20-40 min)
Allows us to visualise disorder such as epilepsy
Epileptic seizures can cause activity with clear abnormalities on standard EEG
-> Extreme form of synchronous activity which is localised or general
EEG used to be used to diagnose coma and brain death
-> Anatomical imaging techniques are now preferred
Voltage and frequency of EEG -
Records small voltage fluctuation between any pair of electrodes
Voltage patterns are recorded in real time - EEG voltage can range from approximately -100mV to +100mV with a frequency of up to 40Hz
Sensorimotor rhythms -
Brain wave rhythms - oscillatory idle rhythms over the sensorimotor cortex (13-15 Hz)
Produced when corresponding sensorimotor areas are idle/immobile
Decrease in amplitude when corresponding sensory or motor areas are activated (during action task or even motor imagery)
Also known as the mu rhythm -
MU rhythms - sensory motor neurons
-> In the part of the brain that controls voluntary movement - motor cortex
-> The rhythm is present when a person is at rest and becomes suppressed when they preform or imagine a motor action
-> Rhythm is blocked by movement execution but also by the observation of other’s actions
Some use Mu waves to investigate its role in learning during child development
Believe it is a marker of an infant’s ability to imitate
Some believe we can learn about autism because they considered it to be an altered mirror neuron system
Mirror neurons in the premotor cortex fire both when a person produces or observes an action
First EEG study investigating mu rhythm in autism (Oberman et al, 2005)
10 autistic participants and 10 neurotypical participants
EEG was measured while children observed hand opening and closing
Mirror neurons in the premotor cortex fire both when a person produces or observes an action
Mu waves of typical children were inhibited when doing and observing
Mu waves of autistic children were inhibited when doing but not observing
Interpreted as - mirror neuron systems in autistic children is different
ERPs
Event related potential =
We time lock the recording of EEG to the onset of an event that we want to investigate -> codes are sent from the stimulus presentation computer directly to the EEG acquisition computer, which reveals a signal which is linked to real-time information processing in the brain
This activity is called an event related potential
How do we derive them?
Time span during which the EEG is time-locked to an event is called an epoch
An epoch generally begins at a specified time before the event occurs (e.g. 100 ms) and persists until some time after the event
Epoch lengths are generally task specific and based on the stage of neural processing of interest during stimulus processing
Codes are stimulus-specific markers that identify the onset of the stimuli presented to the subject
Codes are later used to isolate and group epochs that are time-locked to the same types of stimuli under the same experimental conditions
Each experimental block is composed of many individual trials
Epochs and averaging -
Averaged waveform reflects the average neural activity across time to the specific stimulus
The more trials that contribute to an experimental condition, the greater the signal to noise ratio
Neural activity that is not time locked to the task specific event (background noise) will wash out
Many epochs with the same code get averaged together - an ERP is the averaged brain response that is extracted from multiple epochs that are time-locked to the same type of event or stimulus
Features of an ERP waveform -
Characteristic peaks and troughs that appear during cognitive tasks
These ERP waveform features can be described as components
Components can be characterised by their polarity and latency - some scales differ
Peaks are generally labelled with a P (positive) or N (negative) and are measured in milliseconds from stimulus onset
Early latency components -
Tied to sensory perception / processing
Are often called sensory evoked potentials or visual evoked potentials in the case of visual modality
Associated with physical properties of visual stimuli e.g. spatial orientation, motion and colour)
Are called exogenous components because they are outside voluntary control and independent of the cognitive state of the subject
Later onset components -
These are associated with later and higher cognitive functions or response-related processing occurring with sensory processing
They have unique properties specific to the type of processing required by the event
They can be modulated by the cognitive aspects of a stimulus
Types of measurement -
Measure the distribution of voltage over the scalp
Measure the size and latency of component peaks and troughs
Evaluation -
Non-invasive
Excellent temporal resolution
Poor spatial resolution
Brief changes in EEG signal
Time-aligned with stimulus onset
Averaged over 100s of responses