Magnetic resonance imaging: Current and emerging applications in the study of the central nervous systemDA Sanghvi1, Z Patel2, T Patankar3
1 Department of Radiology, Kokilaben Dhirubhai Ambani Hospital, Andheri West, Mumbai, Maharashtra, India
2 Department of Radiology, PD Hinduja Hospital and Research Center, Mumbai, Maharashtra, India
3 Department of Radiology, Royal Preston Hospital, Lancashire Teaching Hospitals, NHS Foundation Trust, United Kingdom
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0022-3859.65283
Source of Support: None, Conflict of Interest: None
Neuroimaging is presently utilised in clinical practice for initial diagnosis and mapping of disease extent and distribution, noninvasive, preoperative grading of tumours, biopsy planning, surgery and radiation portal planning for tumors, judging response to therapy and finally, prognostication. Newer advances include magnetic resonance (MR) diffusion and diffusion tensor imaging with tractography, perfusion imaging, MR spectroscopy and functional imaging using the blood oxygen level-dependent contrast technique. Neuroimaging plays a pivotal role in various degenerative and neoplastic diseases, improving diagnostic accuracy, affecting patient care, monitoring dynamic changes within the brain during therapy, and establishing them as the arbiter of novel therapy that may one day prove cure of various brain diseases a reality.
Keywords: BOLD, diffusion, perfusion, spectroscopy
Over the past few decades, as novel therapies for patients with neurological and neurosurgical disorders are being developed, we are witnessing a shift in imaging from merely providing anatomical information towards providing information about neurophysiology. Neuroimaging is presently utilised in clinical practice for initial diagnosis and mapping of disease extent and distribution, non invasive, preoperative grading of tumours, biopsy planning, surgery and radiation portal planning for tumors, judging response to therapy and finally, prognostication. Newer advances include magnetic resonance (MR) diffusion and diffusion tensor imaging (DTI) with tractography, perfusion imaging, MR spectroscopy and functional imaging using the blood oxygen level-dependent contrast (BOLD) technique.
Diffusion-weighted MR imaging is the simplest form of diffusion imaging.  It is a pulse sequence sensitised to the random motion of water molecules (which is termed Brownian motion More Details). Certain pathologies constrain the normal random motion of water molecules in brain tissue and this is referred to as restricted diffusion. Diffusion weighting enables one to distinguish between rapid diffusion of protons (unrestricted diffusion) and slow diffusion of protons (restricted diffusion). Lesions that have restricted diffusion appear hyperintense on diffusion images and hypointense on the accompanying apparent diffusion coefficient (ADC) maps. Using an ADC map it is possible to quantify the diffusion in brain tissues.
A more sophisticated extension of diffusion imaging is diffusion tensor imaging. Diffusion tensor MR imaging is a noninvasive in vivo method for mapping white matter fiber tract trajectories in the human brain.  Diffusion tensor imaging is based on the concepts of isotropic and anisotropic diffusion. The movement of water molecules occurs in all three directions.When water molecules diffuse equally in all three directions, this is termed isotropic diffusion. This is typical in the ventricles, but is also true in the grey matter. In the white matter, free water molecules move anisotropically, i.e. water diffusion is not equal in all three directions. This is because in white matter tracts, the myelin sheath surrounding white matter causes water molecules to move more along the long axis of a fiber bundle and less perpendicularly. Maximum diffusivity coincides with the white matter fiber tract orientation.
Information from DTI is presented in two formats which are FA (fractional anisotropy) maps and tractography. FA maps are cross-sectional images that may be in a grey scale format or may be color-coded for directional information. In FA maps structures that have anisotropy i.e. white matter appear bright on grey-scale FA maps and the degree of brightness is proportional to the anisotropy. When a white matter tract is destroyed by say a tumor there is loss of anisotropy and therefore a reduction in the FA values which is manifested on grey-scale FA maps as loss of brightness. FA values can also be quantified numerically. The color FA maps show the direction of white matter tracts. Again the intensity of the color hues is proportional to the extent of anisotropy. In addition to assessment of the diffusion in a single voxel, DTI has been used to map the white matter fiber tracts. These three-dimensional (3D) reconstructions are called tractography. The principle direction of diffusion in a voxel is called the Eigenvector. Tractography is done by connecting a given voxel to the appropriate adjacent voxel in accordance with the direction that the voxels principle eigenvector is oriented .
On imaging the microvascular structure can be studied by high temporal resolution imaging of the passage of the contrast bolus. Following an injection of contrast media the bolus will pass through the vascular bed entering through arterial vessels, passing through the capillaries bed and draining into the venous system. The amount of contrast that passes into the vascular system will depend on the blood flow rate through the vessels and the contrast dose injected. Within any given voxel the amount of intravascular contrast will depend on the proportion of the voxel formed by blood vessels. As contrast passes through the capillary bed it will leak into the extravascular extracellular space (EES). The rate at which this leakage occurs will reflect the difference in contrast concentration between the blood plasma and the EES. For any given concentration ratio the amount of contrast that leaks will also be restricted by the permeability and surface area of the endothelial membrane. Dynamic MR perfusion can be performed using either T2, T2*or T1-weighted images. It can be seen that the behaviour of contrast material within any given voxel will be related to the concentration time course of contrast entering the arterial vessels, the regional blood flow, the local blood volume, and the endothelial permeability, the surface area of the endothelium and the size of the EES. Analysis schemes for dynamic contrast enhanced-MRI are designed to identify biomarkers which represent one of, or combinations of these biological features. Arterial spin labelling is a new perfusion technique that does not require exogenous contrast; instead it exploits the spins of endogenous water protons that perfuse the imaging plane.
MR spectroscopy (MRS) is the only noninvasive technique capable of measuring chemicals within the body. MRS distinguishes various metabolites on the basis of their slightly different chemical shifts or resonance frequencies. Biologically, relevant nuclei that are amenable to MR analysis are those with an odd number of protons and neutrons such as 1 H , 31 P , 13 C , 19 F and 23 Na. Of these, the one we commonly use is hydrogen or proton spectroscopy. The metabolic information received is displayed as a graph. On the X axis are plotted the resonance frequencies which allow us to identify each unique metabolite. These frequencies are plotted in a unit called parts per million (ppm). Using the Y axis, it is possible to quantify the metabolite by either measuring the peak value referred to as the amplitude or the area under the curve, called the integral value. Various ratios are often calculated. Single or multiple voxels of the brain can be interrogated with this technique. Multivoxel spectroscopy is also called chemical shift imaging or CSI. Using CSI it is possible to create visually appealing colour maps or metabolite maps for spatial demonstration of the metabolite peaks and ratios. These colour maps are overlapped or fused with conventional MR techniques to improve anatomical localization.
Functional MRI (F MRI) refers to the demonstration of brain function with neuroanatomic localization on a real-time basis. The vast majority of these studies are performed using BOLD techniquewhich requires the detection of very small signal intensity changes- 0-3% at 1.5 Tesla and up to 6% at 3 Tesla for voxel volumes as small as 3 ΄ 3 ΄ 5 mm.  The principle of the BOLD technique of F MRI is that performing a predefined cognitive task leads to regionally increased neuronal activity and localized hemodynamic changes that produce a signal response.
Stroke remains a leading cause of mortality and morbidity in India and the world. The past decade has witnessed dramatic advances in stroke neuroimaging and therapeutics that includes diffusion-perfusion imaging, multidetector computed tomography (CT) angiography and endovascular clot dissolution/retrieval techniques.
Most of these advances have been in the imaging and management of hyperacute ischemic stroke. ,, The primary aim of imaging in hyperacute ischemic stroke is direct management in terms of salvaging the penumbra. Reversal of penumbra is associated with a significant decrease in morbidity and mortality in stroke. Currently, MR perfusion studies in combination with diffusion images are used in the setting of hyperacute ischemic stroke to establish the presence of a penumbra which is an indication for thrombolytic therapy in the first 6 h after the onset of an acute neurologic deficit. Penumbra is the tissue at risk for dying if ischemia continues without recanalization of an intravascular thrombus. The abnormality seen on a diffusion image represents the infarct core where damage is irreversible. The ischemic penumbra is at the area around the infarcted core which may be salvaged by thrombolysis. It is depicted on colour perfusion maps as an area around with the infarct core with elevated mean transit time (MTT), decreased cerebral blood volume (CBV) and decreased cerebral blood flow (CBF). The absence of a significant diffusion- perfusion mismatch implies that the patient will not benefit from thrombolysis. A large diffusion abnormality suggestive of a large infarct associated with a small perfusion abnormality representing a small penumbra should be treated with caution, taking into consideration other factors such as time after onset of symptoms. This is because of a potential increase in the risk of subsequent hemorrhage that is associated with an initial large infarct size. Finally, a small diffusion abnormality representing a small infarct with a large perfusion defect representing a significant penumbra [Figure 1] is an indication for thrombolysis in the absence of hemorrhage. Follow-up imaging of patients with untreated or unsuccessfully treated diffusion-perfusion mismatch exhibit substantial lesion growth on follow-up imaging. On the other hand, those with early complete recanalization do not exhibit lesion progression on follow-up imaging. These observations emphasize the accuracy of diffusion-perfusion imaging in guiding management and in prognostication of patients with acute ischemic stroke.
In the future, selection of patients for thrombolytic therapy may be made effective by performing appropriate imaging studies rather than relying on time of onset as the sole determinant of selection.  In recent trials, , intravenous desmoteplase injection at 3-9 h after onset was associated with a higher rate of reperfusion and better clinical outcome in patients selected because of a mismatch between the findings of diffusion and perfusion MR images. The symptomatic intracranial hemorrhage rate was also low.
In addition to its critical role in the management of hyperacute stroke, perfusion may also be used to assess stroke risk by measurement of cerebrovascular reserve capacity.  The ability of the brain to adapt to sudden onset ischemia and hypoxia by autoregulatory vasodilatation is termed cerebrovascular reserve capacity. In patients with chronic cerebral ischemia due to carotid disease, this autoregulatory mechanism is often exhausted due to long-term lack of blood supply and therefore this subset of patients are at increased risk of significant damage if an acute ischemic event occurs. MR and CT perfusion can demonstrate lack of response in cases of chronic carotid stenosis when performed before and after the administration of an intravenous pharmaceutical that induces vasodilatation such as acetazolamide. Lack of increase in cerebral perfusion after administration of an exogenous vasodilator implies lack of cerebrovascular reserve capacity and increased risk during hemodynamic stress.
Finally, a number of recent studies , have highlighted the ability of perfusion MR for evaluating postoperative alteration of cerebral hemodynamics following superficial temporal artery - middle cerebral artery (STA-MCA) anastomosis in patients with Moya Moya disease. These studies have shown that shortening of the time to peak (TTP) in the MCA territory of the hemisphere operated on is a marker of the development of collateral circulation form the external carotid artery to the internal carotid artery.
Epilepsy is a common disorder, with a prevalence of 0.4-1% of the population . ,, Apart from being a chronic medical illness, its ill effects affect social and economic life. Imaging plays a role in identifying structural abnormalities which require surgery and in diagnosing a specific syndrome or etiology. MRI is the best for depicting structural abnormality, while nuclear medicine and specialized MR techniques give additional functional information. 
CT is now usually performed only in the setting of an emergency and trauma associated with focal neurological deficits. It diagnoses large lesions like tumors, granulomas and vascular malformations. MRI has its distinct advantages due to its multiplanar capability, higher spatial resolution, excellent soft tissue contrast and lack of ionizing radiation.  The MRI protocols are tailored depending on the type of seizure. In Temporal lobe epilepsy (TLE), oblique thin coronal T2, FLAIR (fluid attenuated inversion recovery) and T1-weighted gradient volume sequences (SPGR or MP-RAGE) are obtained through the hippocampus. MR findings in histologically proved cases of mesial temporal sclerosis (MTS) are hippocampal atrophy and increased T2 signal  [Figure 2]. The 3D SPGR sequence has revolutionized imaging of extratemporal lesions (cortical dysplasia, heterotopias, polymicrogyria etc), its advantage being its excellent spatial resolution and the ability to reformat. Contrast studies are mainly reserved for focal mass lesions, adult onset seizures or a few syndromes. Using a surface coil MRI and high-strength scanners can also aid in locating cortical dysplasia. ,
In cases of bilateral MTS, one side is affected more than the other in 80% of patients.  Quantitative techniques have proved to be useful since they can assess bilaterality and grade the disease. Hippocampal volumetry is one of the MR methods to study the hippocampus quantitatively. , The area of the hippocampus is manually measured on each image on the 3D SPGR sequence. A significant difference between the volumes on both sides is a reliable pointer of unilateral hippocampal sclerosis. When the involvement is bilateral the measurements are compared with reference values for the normal population. T2 relaxometry quantifies the T2 hydration.  A series of T2W images of different TE are taken through the same slice and the T2 decay is measured. Normally it is less than 110-115 msec. In MTS the T2 time is elevated, and abnormal T2 relaxometry is significantly associated with intractable epilepsy.  Proton 1 H spectroscopy as stated earlier maps the various brain metabolites. N-acetyl-aspartate (NAA) is a neuronal marker. Decreased values of NAA/creatinine and NAA/creatinine+choline ratios imply neuronal loss and/or metabolic dysfunction. Decreased ratios have been able to lateralize TLE in 65-90% of patients with bilateral temporal lobe structural abnormalities.  In cases of temporal lobe epilepsy (TLE) with normal MR studies, NAA ratios can provide lateralizing evidence in at least 20% of patients.  31 P spectroscopy when done during and after a seizure depicts changes in the pH and high-energy phosphates.  Its role in TLE is still controversial.
Functional techniques, including positron emission tomography (PET), single-photon emission computerized tomography (SPECT), and fMRI are useful tools for localizing the epileptogenic zone and mapping functional areas of the brain such as language and motor function. Epilepsy surgery usually involves a resection or disconnection technique which removes or isolates the epileptogenic zone, while sparing functional cortex.  fMRI is used for mapping the eloquent cortex before surgery and denote its relationship with the epileptogenic zone. It is particularly useful for children, since there is no radiation exposure, and it is easy to repeat if the test is inconclusive.  Presurgical [Figure 3] and postsurgical DTI with 3D tractography is done on some occasions for evaluating the completeness of the surgery and complications in case of new post-surgical deficits. SPECT using blood flow markers such as HMPAO reveals a region of relative hypoperfusion interictally and an ictal focal increase in perfusion. Fluro deoxy glucose (FDG) PET has been more valuable in assessing severe epilepsy in very young children, where it reveals focal hypometabolism amenable to highly successful surgical resection.  A comparison of MRI, PET and ictal SPECT, using pathological diagnosis as a standard of reference, showed correct lateralization in 72%, 85% and 73%, respectively. 
In spite of the wide range of available MR techniques, there still is a group of MR imaging-negative patients. In these cases novel techniques like diffusion, magnetization transfer imaging (MTI) and T2 mapping and double inversion recovery which identify microstructural abnormalities can be attempted. Decreased anisotropy and increased ADC values are seen in the sclerosed hippocampi in the interictal phase.  Whole brain T2 mapping and MTI quantify the T2 and MTR values respectively and corroborate well with the abnormality on conventional images but also help in identifying occult lesions.  However, the there is still limited data available and at present their role is mainly for guiding the accurate placement of intracranial electro-encephalogram (EEG) electrodes. 
Primary malignant brain tumours are associated with the third highest cancer-related mortality rate with a disproportionate level of disability and morbidity. The current WHO classification, however, falls short of predicting the therapeutic response of each individual tumour within the same histological grade and cannot provide precise guidance of therapy, especially those targeting specific molecular or genetic pathways of tumour genesis . , In most cases the radiological diagnosis of brain tumour is obvious but if the nature of the lesion is still in question after comprehensive investigation, further imaging with advanced technique such as diffusion [Figure 4], perfusion or spectroscopy may be warranted to grade tumors, assess cellularity and differentiate brain cancers from tumour-mimicking lesions such as infarcts, abscesses or demyelination. ,,
The analysis of dynamic contrast-enhanced MRI data using pharmacokinetic models which describe the behaviour of contrast within the tissue has enabled the development of specific biological surrogate markers including flow [F], endothelial permeability surface area product [K trans ], blood volume [CBV] and extracellular extravascular tissue volume [Ve].
Dynamic contrast-enhanced MR is being used to assess the vascularity of intracranial lesions in gliomas, ,, tumour-mimicking demyelination  and cerebral lymphomas.  Increased tumour vascularity is not synonymous with malignancy. There are several intracranial neoplasms; especially those that are extra-axial in location such as meningiomas or choroid plexus papillomas, can be highly vascular but rather benign in biologic behaviour. Differentiation between radiation necrosis and recurrent tumour may be difficult on conventional imaging, however, dynamic contrast-enhanced MR imaging may be able to differentiate as radiation necrosis shows low CBV and K trans . ,,,,, Meningiomas are highly vascular extra tumours that develop blood supply from meningeal arteries with tumour capillaries that completely lack blood-brain barriers. They appear hypervascular on perfusion MR and the capillaries are very leaky and permeable and therefore the CBV measurements may be grossly overestimated or under-estimated on T2* imaging. Multiple intracranial metastatic diseases often do not pose a diagnostic dilemma but a solitary metastatic brain tumour can appear similar to glioma on post-contrast T1 images and CBV maps. However, it has been shown that the perfusion MR may be useful in differentiating solitary metastases and primary glioma based on the difference in the peritumour CBV measurements.  This can be explained by the fact that the peritumour edema is pure vasogenic edema in metastatic tumours suggesting no tumour beyond the contrast-enhancing margin in metastases. In high-grade tumours however the perilesional edema may represent a combination of vasogenic edema and tumour infiltration around the perivascular spaces.  Zhu et al., calculated K trans and v e maps in patients with meningioma, glioma and acoustic neuroma. K trans in acoustic neuromas was found to be consistently lower than that observed in meningioma and glioma but v e (size of EES) was significantly higher in acoustic neuroma and also high in meningioma compared to glioma. Primary cerebral lymphoma may mimic demyelinating lesions or tumours but show relative decreases in CBV on perfusion MR may be useful in differentiating between the tumour types which are managed and treated differently. , Tumefactive demyelinating lesions may sometimes be difficult to differentiate from primary brain tumours as they may develop a mildly elevated rCBV and characteristic intra-lesion venous enhancement but marked hypovascularity is not seen. 
Several studies ,,, have shown a statistically strong correlation between tumour rCBV and astrocytoma grading, [Figure 5] as well as conventional angiographic tumour vascularity. Low-grade astrocytomas have significantly lower average CBV than an anaplastic astrocytoma or glioblastoma. ,, Astrocytoma grading using rCBV, must be limited to fibrillary astrocytomas because other gliomas, most notably oligodendrogliomas, may have high rCBV regardless of grade. 
The relationship between K trans and grade remains less clear ,,, [Figure 3]. We have demonstrated strong relationships between both CBV and K trans and histological grade in gliomas and either measurement, or a combination of the two, show good discriminative power in distinguishing between low- and high-grade tumours.  [Figure 6]. Histological grade, patient age, use of adjuvant therapy, degree of surgery, and size of tumour have all also shown, to a lesser extent to have some prognostic value. ,,,6,,, Mills et al., have shown that histological grade remains the most statistically significant predictor of prognosis when compared to CBV and K trans with CBV simply a reflection tumour grade and has no additional prognostic value if histological grading is performed. But K trans demonstrates a statistically significant relationship with prognosis, which is independent of tumour grade.
Neurodegenerative processes including Alzheimer's disease (AD), frontotemporal dementia (FTD) and dementia with Lewy bodies (LBD) can each cause severe dementing syndromes. However, in patients with a combination of neurodegenerative and vascular abnormalities the relative significance of the separate pathological processes to the clinical features of the disease remains unclear.
Ischemic-vascular dementia (IVD) is commonly associated with widespread small ischemic or vascular lesions throughout the brain with predominant involvement of the basal ganglia, white matter, and hippocampus.  Several groups have shown that severe lacunar state and micro infarction due to arteriolosclerosis and hypertensive microangiopathy are more common in IVD subjects than in normals and have emphasized the importance of small vascular lesions in the development of dementia. ,,, Many groups have suggested that simple scoring schemes of white matter lesion load and distribution are useful in the diagnosis of vascular dementia. ,,, However, although white matter lesions are more severe in patients with vascular dementia, ,,,,,,,, they are more prevalent in all dementia groups than in normals.
It is widely considered that AD is distinguishable from normal brains and those affected by vascular dementia (VaD) or frontal lobe dementia from images showing bilateral temporoparietal hypoperfusion or hypometabolism (impaired metabolism).
It has also long been known that AD is more severe in the presence of established cerebral infarction , and autopsy studies show 70% of patients with definite AD to have evidence of coexistent cerebrovascular disease and 35% to have had a cerebral infarction. ,,, Structural imaging with CT and MRI only revealed mild cortical atrophy in AD but numerous small infarcts in VaD. Oxygen extraction fraction [OEF], especially in the parietotemporal cortex, was found to be raised in probable AD but not in VaD, whereas vascular transit time [VTT] and vascular reactivity [VR] were normal in AD but the VaD patient had prolonged VTT and decreased VR, CBF and cerebral metabolic rate of oxygen (CMRO2). 
Combining findings from different modalities, such as atrophy of the medial temporal lobe, as measured with CT, with parietotemporal hypoperfusion as measured with SPECT, can apparently improve diagnostic accuracy for AD, as this combination is much less common in other dementias and rare in normal brains. 
The term "mixed dementia" is often used to describe patients with neuropsychological deficits in keeping with AD but with clear evidence of coexistent vascular disease fulfilling the National Institute of Neurological Disorders and Stroke and Association Internationale pour la Recherche er l'Enseignement en Neurosciences (NINDS-AIREN) criteria for the diagnosis of vascular dementia .  It has been proposed that coexistent cerebrovascular disease and AD may have a synergistic effect and thus patients present earlier and or have more cognitive impairment than would be expected if either disease were to present in isolation. ,,,, The CATCH hypothesis put forward by de la Torre proposes that in patients in the early stages of AD or in those at high risk of developing the disorder, the presence of an ischemic vascular insult producing a "critically attained threshold of cerebral hypoperfusion" (CATCH) promotes or triggers the clinical onset of AD. ,,,
Virchow-Robin spaces are microscopic perivascular spaces, which surround the perforating arteries that enter the brain but, when dilated they may be seen on MR examinations. We recently demonstrated that VRS dilatation is commoner in diseases associated with microvascular angiopathy data on the [Figure 7]. Abnormal VRS dilatation is shown to be associated with vascular dementia and with treatment resistance in late-onset depression (LOD).  A recent study demonstrated that SCE measured by continuous Transcranial Doppler (TCD) of the middle cerebral artery and venous to arterial circulation shunt (v-aCS) indicative of patent foramen ovale (PFO) were commoner in both AD and VaD. ,, A study using quantitative [1H]MRS found that patients with a diagnosis of 'probable AD' had 20% less cerebral (glutamate + glutamine) than normal.  The Moats team also found a 50% increase in myoinositol in the cognitively-impaired sample, consistent with the 2002 MRS findings by Waldman et al., of a higher myoinositol:creatine ratio in AD patients compared with controls and patients with vascular dementia. 
Neuroimaging plays a pivotal role in various degenerative and neoplastic diseases, improving diagnostic accuracy, affecting patient care, monitoring dynamic changes within brain during therapy, and establishing them as the arbiter of novel therapy that may one day prove cure of various brain diseases a reality.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7]