Lecture 6: Multi-modal Imaging (Neuroimaging) Flashcards
- LEARNING OUTCOMES
Neuroimaging Computing Methods and Applications
- Overview of Medical Image Computing
- Imaging Modalities and Methods
- Multi-modal Imaging Systems
- Applications
- List some of the Imaging Modalities and Methods used in Neuroimaging
Tomography: CT, PET, SPECT, etc.
Electrography: EEG, MEG (Magnetoencephalography)
Clinical MRI: T1, T2, PD, FLAIR, T1-Enhanced
Research MRI: DTI, fMRI, MTR, SWI, ASL
Multi-modal Imaging Systems: PET-CT, PET-MRI, EEG-MRI, MRI-MRI
- List some of the applications of MM MIC.
Research Projects:
- Neurodegeneration Prediction
- Human Brrain Connectome
Clinical Applications:
- White Matter Lesion Identification
- Neurosurgery Planning
- Briefly describe Medical Image Computing (MIC); goal and focus.
The continued evolution of complexity and info richness of multidimensional datasets in MI has spurred a parallel development of increasingly sophisticated and capable medical image computing (MIC) platforms for medical research as well as clinical care.
The main goal of MIC is to extract clinically relevant info or knowledge from medical images.
While closely related to the field of MI, MIC focuses on the computational analysis of the images, not their acquisition.
- The impact of advanced MI today is the result of 3 concurrent, interlinked revolutionary technological developments, what are they?
Data Acquisition - anato**mically correlated digital data acquisition from the body
Image Processing - advanced multistage image processing methods to extract and enhance information of relevance for research and clinical use
Platform Development - the continuing evolution of advanced computer science techniques to organise and sustain high-quality platforms for data processing
- Describe Computed Tomography (CT) and why there is a limitation to its use.
CT uses ionising X-rays to take images from different angles in very fine slices through the specific part of the body.
It uses a back projection scheme (?) to combine these X-ray images to produce the slices (tomographic images) of the body.
Because of the risks of ionising radiation associated with CT scans, patients should not receive CT screening in excess of those recommended by established guidelines.
- How might Positron Emission Tomography (PET) scans be useful in brain imaging?
PET is the most powerful and versatile approach to study neurotransmitter/receptor interactions.
PET is inherently a molecular imaging technique, exquisitely sensitive for detecting targeted molecules - targeted by choice of radiotracers.
- Eg a radiotagged glucose, FDG, was a widely used radiotracer that can assess glucose metabolism in the brain. This has made it useful for diagnosis, staging, and monitoring treatment of cancers
Recently, however, FDG-PET studies have declined with new tracers introduced.
- Eg amyloid-binding compounds have been reported as probes for imaging amyloid plaques in Alzheimer’s brains
- What do EEG and MEG stand for and how are they useful in imaging the brain?
What are their shortcomings?
EEG - Electroencephalography
MEG - Magnetoencephalography
EEG - electrophysiological monitoring method, records brain electrical activity. Typically non-invasive - electrodes placed along scalp - measures voltage fluctuations resulting from the ionic current within neurons.
MEG - functional technique - maps brain activity by recording magnetic fields produced by naturally occurring brain currents using sensitive magnetometers.
Clinically - EEG = recording spontaneous electrical activity over period of time (as recorded from electrodes on scalp)
Recent advances in EEG/MEG have improved localisation of event-related brain activity and of intracranial spikes in epilepsy patients.
Both detect activity below the cortex poorly. Their respective errors however are very different, and combining them may allow for some noise corrections.
Opportunity: high temporal resolution EEG/MEG source imaging with high spatial resolution fMRI. Much progress has been made to leverage the complementary nature of EEG & fMRI, which can be performed simultaneously in an MRI scanner.
- Discuss briefly regarding MRI:
- What it is
- How it works
- Benefits
- Limitations
What - imaging technique primarily for medical settings to create HQ images of human body based on principles of nuclear magnetic resonance (NMR)
How
- MRI creates a steady state of magnetism within body
- stimulates body with radio waves
- stops the radio waves and registers body’s electromagnetic transmission
- transmitted signals used to construct the internal images of the body
Benefits
- safe - uses magnetic and radio waves to generate images - not ionising radiation
- no harmful side-effects associated with temporary exposure to strong magnetic field (8 Tesla) and radio waves used by MRI scanners
Limitations
- potential interactions with implanted devices eg pacemakers, cardioverter-defibrillators (ICDs) - patients with these are restricted from MRI examinations by FDA
- Explain T1-weighted MRI scans and their use in MI.
T1 time of a tissue is the time it takes for excited spins to recover and be available again for the next excitation.
-
short T1 tissues appear bright - regain most longitudinal magnetisation during the TR (repetition time) interval = produce stronger MR signal, eg fat
- short = bright = stronger signal
-
long T1 tissues appear dark - don’t regain much longitudinal magnetisation during TR interval = proudce a weaker MR signal, eg CSF
- long = dark = weaker signal
- T1-weighted MRIs may be acquired after injection of a contrast agent - gadolinium. Why is this done and how is it useful?
Enables analysis of blood vessels generated by brain tumours/lesions.
The blood vessels and pathologies with high vascularity appear bright on T1 weighted post-gadolinium images.
- Explain T2-weighted MRI scans and their use in MI.
The T2 time determines how quickly an MR signal fades after excitation.
- short T2 tissues appear dark, eg Fat
- long T2 tissues appear bright, eg CSF
Pathological processes normally increase the water content in tissues, therefore they usually appear bright on T2-weighted images.
- How does Proton Density Weighted MRI (PD) work?
The proton density (PD) is the number of excitable spins per unit volume - which determines the maximum signal that can be obtained from a given tissue.
The image contrast is not dependant on T1/T2 relaxation - signal received is dependent on the amount of protons in the tissues
- Less protons - low signal - dark areas
- more protons - more signal - bright areas
Proton density can be enhanced by minimising the effect of T1 and T2 contrast.
PD sequences useful for evaluating structures with low signal intensities - bone, connective tissue (ligaments, tendons) - mainly used for imaging of brain, spine, musculoskeletal system
- What does FLAIR stand for and how is it used in MRIs?
FLAIR = Fluid-Attenuated Inversion Recovery
A pulse sequence used in MRI that nulls fluids, thus is used in CNS where lesions are normally covered by bright CSF signals.
Like T2 - pathological processes increase water content in tissues, which results in signal increase on FLAIR images, thus they appear brighter on FLAIR images also.
- What brain technique allows us to infer the white matter pathways of the brain in vivo? Discuss.
Diffusion MRI (dMRI) aka diffusion tensor imaging (DTI)
dMRI most applicable when tissue of interest is dominated by isotropic fluid movement, eg in brain gm or CSF. By probing at different orientations, dMRI estimates the orientation of axonal fibre bundles based on the fact that water diffuses most rapidly along the length of axons. This leads to longer scanning time compared to sMRI (structural MRI?).
dMRI is used experimentally and not yet evaluated in clinical trials due to differences in signal estimation models and fibre tracking algorithms, variations in datasets and lack of ground truth. Despite this - rapid development - new models and methods every year. Large-scale datasets with uniformly collected dMRI data also growing in size which will facilitate the evaluation of these models and methods.
