Scanning Conditions in Functional Connectivity Magnetic Resonance Imaging: How to Standardise Resting-State for Optimal Data Acquisition and Visualisation?

Functional connectivity magnetic resonance imaging (fcMRI), performed during resting wakefulness without tasks or stimulation, is a non-invasive technique to assess and visualise functional brain networks in vivo. Acquisition of resting-state imaging data has become increasingly common in longitudinal studies to investigate brain health and disease. However, the scanning protocols vary considerably across different institutions creating challenges for comparability especially for the interpretation of findings in patient cohorts and establishment of diagnostic or prognostic imaging biomarkers. The aim of this chapter is to discuss the effect of two experimental conditions (i.e. a low cognitive demand paradigm and a pure resting-state fcMRI) on the reproducibility of brain networks between a baseline and a follow-up session, 30 (±5) days later, acquired from 12 right-handed volunteers (29 ± 5 yrs). A novel method was developed and used for a direct statistical comparison of the test-retest reliability using 28 well-established functional brain networks. Overall, both scanning conditions produced good levels of test-retest reliability. While the pure resting-state condition showed higher test-retest reliability for 18 of the 28 analysed networks, the low cognitive demand paradigm produced higher test-retest reliability for 8 of the 28 brain networks (i.e. visual, sensorimotor and frontal areas); in 2 of the 28 brain networks no significant changes could be detected. These results are relevant to planning of longitudinal studies, as higher test-retest reliability generally increases statistical power. This work also makes an important contribution to neuroimaging where optimising fcMRI experimental scanning conditions, and hence data visualisation of brain function, remains an on-going topic of interest. In this chapter, we provide a full methodological explanation of the two paradigms and our analysis so that readers can apply them to their own scanning protocols.

Brain hyperintensities in magnetic resonance imaging of patients with mild acute focal neurology.

PURPOSE: Hyperintensities are common in neuroimaging scans of patients with mild acute focal neurology. However, their pathogenic role and clinical significance is not well understood. We assessed whether there was an association between hyperintensity score with diagnostic category and clinical assessments/measures. METHODS: One hundred patients (51 ± 12 years; 45:55 women:men), with symptomatology suggestive of short duration ischemia referred for magnetic resonance imaging, were prospectively recruited in NHS Grampian between 2012 and 2014. Hyperintensities were quantified, on T2 and FLAIR, using the Scheltens score. RESULTS: The most frequent diagnosis was minor stroke (33%), migraine (25%) and transient ischemic attack (17%). The mean total Scheltens score was 28.49 ± 11.93 with all participants having various loads of hyperintensities. Statistically significant correlations between hyperintensity scores and clinical assessments/measures (age, systolic blood pressure, pulse pressure, MoCA) at the global level were also reflected regionally. These provide further supporting data in terms of the robustness of the Scheltens scale. CONCLUSION: Hyperintensities could serve as a diagnostic and prognostic imaging biomarker for patients, presenting with mild acute focal neurology, warranting application of automated quantification methods. However, larger cohorts are required to provide a definitive answer especially as this is a heterogenous group of patients.

Cerebrovascular reactivity measurement in cerebral small vessel disease: Rationale and reproducibility of a protocol for MRI acquisition and image processing.

Background Impaired autoregulation may contribute to the pathogenesis of cerebral small vessel disease. Reliable protocols for measuring microvascular reactivity are required to test this hypothesis and for providing secondary endpoints in clinical trials. Aims To develop and assess a protocol for acquisition and processing of cerebrovascular reactivity by MRI, in subcortical tissue of patients with small vessel disease and minor stroke. Methods We recruited 15 healthy volunteers, testing paradigms using 1- and 3-min 6% CO2 challenges with repeat scanning, and 15 patients with history of minor stroke. We developed a protocol to measure cerebrovascular reactivity and delay times, assessing tolerability and reproducibility in grey and white matter areas. Results The 3-min paradigm yielded more reproducible data than the 1-min paradigm (CV respectively: 7.9-15.4% and 11.7-70.2% for cerebrovascular reactivity in grey matter), and was less reproducible in white matter (16.1-24.4% and 27.5-141.0%). Tolerability was similar for the two paradigms, but mean cerebrovascular reactivity and cerebrovascular reactivity delay were significantly higher for the 3-min paradigm in most regions. Patient tolerability was high with no evidence of greater failure rate (1/15 patients vs. 2/15 volunteers withdrew at the first visit). Grey matter cerebrovascular reactivity was lower in patients than in volunteers (0.110-0.234 vs. 0.172-0.313%/mmHg; p < 0.05 in 6/8 regions), as was the white matter cerebrovascular reactivity delay (16.2-43.9 vs. 31.1-47.9 s; p < 0.05 in 4/8 regions). Conclusions An effective and well-tolerated protocol for measurement of cerebrovascular reactivity was developed for use in ongoing and future trials to investigate small vessel disease pathophysiology and to measure treatment effects.

Oxygen challenge magnetic resonance imaging in healthy human volunteers.

Oxygen challenge imaging involves transient hyperoxia applied during deoxyhaemoglobin sensitive (T2*-weighted) magnetic resonance imaging and has the potential to detect changes in brain oxygen extraction. In order to develop optimal practical protocols for oxygen challenge imaging, we investigated the influence of oxygen concentration, cerebral blood flow change, pattern of oxygen administration and field strength on T2*-weighted signal. Eight healthy volunteers underwent multi-parametric magnetic resonance imaging including oxygen challenge imaging and arterial spin labelling using two oxygen concentrations (target FiO2 of 100 and 60%) administered consecutively (two-stage challenge) at both 1.5T and 3T. There was a greater signal increase in grey matter compared to white matter during oxygen challenge (p < 0.002 at 3T, P < 0.0001 at 1.5T) and at FiO2 = 100% compared to FiO2 = 60% in grey matter at both field strengths (p < 0.02) and in white matter at 3T only (p = 0.0314). Differences in the magnitude of signal change between 1.5T and 3T did not reach statistical significance. Reduction of T2*-weighted signal to below baseline, after hyperoxia withdrawal, confounded interpretation of two-stage oxygen challenge imaging. Reductions in cerebral blood flow did not obscure the T2*-weighted signal increases. In conclusion, the optimal protocol for further study should utilise target FiO2 = 100% during a single oxygen challenge. Imaging at both 1.5T and 3T is clinically feasible.

Resting state connectivity and cognitive performance in adults with cerebral autosomal-dominant arteriopathy with subcortical infarcts and leukoencephalopathy.

Cognitive impairment is an inevitable feature of cerebral autosomal-dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL), affecting executive function, attention and processing speed from an early stage. Impairment is associated with structural markers such as lacunes, but associations with functional connectivity have not yet been reported. Twenty-two adults with genetically-confirmed CADASIL (11 male; aged 49.8 ± 11.2 years) underwent functional magnetic resonance imaging at rest. Intrinsic attentional/executive networks were identified using group independent components analysis. A linear regression model tested voxel-wise associations between cognitive measures and component spatial maps, and Pearson correlations were performed with mean intra-component connectivity z-scores. Two frontoparietal components were associated with cognitive performance. Voxel-wise analyses showed an association between one component cluster and processing speed (left middle temporal gyrus; peak -48, -18, -14; ZE = 5.65, pFWE corr = 0.001). Mean connectivity in both components correlated with processing speed (r = 0.45, p = 0.043; r = 0.56, p = 0.008). Mean connectivity in one component correlated with faster Trailmaking B minus A time (r = -0.77, p < 0.001) and better executive performance (r = 0.56, p = 0.011). This preliminary study provides evidence for associations between cognitive performance and attentional network connectivity in CADASIL. Functional connectivity may be a useful biomarker of cognitive performance in this population.

Stroke recovery and lesion reduction following acute isolated bilateral ischaemic pontine infarction: a case report.

BACKGROUND: Although pontine strokes account for a small percentage of all ischaemic events, they can be associated with significant initial disability. These lesions may be missed on computed tomography and therefore magnetic resonance imaging is generally preferred for the assessment of brainstem strokes. The aetiopathogenesis of isolated pontine infarcts, not due to a significant compromise (occlusion or dissection) in the vertebrobasilar territory, still remains to be fully characterised. These strokes present with different symptoms, depending on the lesion location and size, partly reflecting the anatomical variability of the vertebrobasilar vessels. Progressive neurological deterioration is relatively common and has been associated with the extension of such lesions. However, many patients with significant infarcts in the pons will do well in the future and initial diffusion-weighted imaging may not add useful prognostication to the clinical assessment. We discuss here a case where an initially progressive presentation was associated with a marked improvement in both clinical and radiological assessments at 42 days. CASE PRESENTATION: A 49-year-old white British man presented with left-sided weakness, incoordination, unsteadiness, cerebellar ataxic dysarthria and dysphonia. A baseline magnetic resonance imaging scan with diffusion-weighted imaging, T1-weighted and T2-weighted sequences showed an acute bilateral pontine infarct. On a repeat scan at 42 days, there was a 57.5% decrease in the size of the lesion on the high-resolution three-dimensional T1-weighted image and a corresponding improvement in the symptoms and the clinical assessments of this patient. The reduction in infarct size was also comparable to the decrease calculated between the baseline diffusion-weighted and the follow-up fluid attenuated inversion recovery sequences. CONCLUSION: This case report discusses the significant clinical improvement and corresponding lesion reduction in a patient that presented with worsening neurological symptoms and was diagnosed with acute bilateral ischaemic pontine infarction. Further studies, utilising structural and functional magnetic resonance imaging with follow-up scans, are needed to provide better insights into the underlying aetiopathology and recovery mechanisms of pontine stroke. These will help define the relationship between imaging parameters and outcome allowing for better prognosis along with the development of relevant rehabilitation programs for this group of patients.

Dietary nitrate reduces skeletal muscle oxygenation response to physical exercise: a quantitative muscle functional MRI study.

Dietary inorganic nitrate supplementation (probably via conversion to nitrite) increases skeletal muscle metabolic efficiency. In addition, it may also cause hypoxia-dependent vasodilation and this has the potential to augment oxygen delivery to exercising skeletal muscle. However, direct evidence for the latter with spatial localization to exercising muscle groups does not exist. We employed quantitative functional MRI (fMRI) to characterize skeletal muscle oxygen utilization and replenishment by assessment of tissue oxygenation maximal change and recovery change, respectively. Eleven healthy subjects were enrolled, of whom 9 (age 33.3 ± 4.4 years, five males) completed the study. Each subject took part in three MRI visits, with dietary nitrate (7cl concentrated beetroot juice) consumed before the third visit. During each visit fMRIs were conducted concurrently with plantar flexion exercise at workloads of 15% and 25% maximum voluntary contraction (MVC). No significant changes were found between visits 1 and 2 in the fMRI measures. A decrease in maximal change was found at 15% MVC in soleus between visits 2 and 3 (5.12 ± 2.36 to 2.55 ± 1.42, P = 0.004) and between visits 1 and 3 (4.43 ± 2.12 to 2.55 ± 1.42, P = 0.043), but not at 25% MVC or within gastrocnemius. There was no difference in recovery change between visits. We found that dietary nitrate supplementation reduces tissue oxygenation alterations during physical exercise in skeletal muscle. This effect is more prominent in muscles with predominantly type 1 fibers and at lower workloads. This indicates that in healthy subjects dietary nitrate predominantly affects skeletal muscle energy efficiency with no change in oxygen delivery.

Cluster size statistic and cluster mass statistic: two novel methods for identifying changes in functional connectivity between groups or conditions.

Functional connectivity has become an increasingly important area of research in recent years. At a typical spatial resolution, approximately 300 million connections link each voxel in the brain with every other. This pattern of connectivity is known as the functional connectome. Connectivity is often compared between experimental groups and conditions. Standard methods used to control the type 1 error rate are likely to be insensitive when comparisons are carried out across the whole connectome, due to the huge number of statistical tests involved. To address this problem, two new cluster based methods--the cluster size statistic (CSS) and cluster mass statistic (CMS)--are introduced to control the family wise error rate across all connectivity values. These methods operate within a statistical framework similar to the cluster based methods used in conventional task based fMRI. Both methods are data driven, permutation based and require minimal statistical assumptions. Here, the performance of each procedure is evaluated in a receiver operator characteristic (ROC) analysis, utilising a simulated dataset. The relative sensitivity of each method is also tested on real data: BOLD (blood oxygen level dependent) fMRI scans were carried out on twelve subjects under normal conditions and during the hypercapnic state (induced through the inhalation of 6% CO2 in 21% O2 and 73%N2). Both CSS and CMS detected significant changes in connectivity between normal and hypercapnic states. A family wise error correction carried out at the individual connection level exhibited no significant changes in connectivity.

Functional connectivity magnetic resonance imaging in stroke: an evidence-based clinical review.

Stroke is a common condition that may lead to various degrees of neurological deficit and long-term disability. It has become increasingly recognized that cortical reorganization of neuronal networks plays a significant role in regaining function following a focal brain injury. However, the mechanisms involved in this process are still not fully understood. Resting-state functional connectivity magnetic resonance imaging is a rapidly evolving scanning technique that has the potential to shed light into this neuronal rearrangement. A better understanding of the underlying neurological pathways may contribute to the development of targeted treatment that will promote repair and reduce poststroke deficit. The aim of this review is to provide an up-to-date summary of the available scientific data evaluating the clinical application of functional connectivity magnetic resonance imaging among stroke survivors.

The impact of latent confounders in directed network analysis in neuroscience.

In the analysis of neuroscience data, the identification of task-related causal relationships between various areas of the brain gives insights about the network of physiological pathways that are active during the task. One increasingly used approach to identify causal connectivity uses the concept of Granger causality that exploits predictability of activity in one region by past activity in other regions of the brain. Owing to the complexity of the data, selecting components for the analysis of causality as a preprocessing step has to be performed. This includes predetermined-and often arbitrary-exclusion of information. Therefore, the system is confounded by latent sources. In this paper, the effect of latent confounders is demonstrated, and paths of influence among three components are studied. While methods for analysing Granger causality are commonly based on linear vector autoregressive models, the effects of latent confounders are expected to be present also in nonlinear systems. Therefore, all analyses are also performed for a simulated nonlinear system and discussed with regard to applications in neuroscience.

Global functional connectivity reveals highly significant differences between the vegetative and the minimally conscious state.

A major challenge in the diagnosis of disorders of consciousness is the differential diagnosis between the vegetative state (VS) and the minimally conscious state (MCS). Clinically, VS is defined by complete unawareness, whereas MCS is defined by the presence of inconsistent but clearly discernible behavioural signs of consciousness. In healthy individuals, pain cries have been reported to elicit functional activation within the pain matrix of the brain, which may be interpreted as empathic reaction. In this study, pain cries were presented to six VS patients, six MCS patients, and 17 age-matched healthy controls. Conventional task-related functional magnetic resonance imaging (fMRI) showed no significant differences in functional activation between the VS and MCS groups. In contrast to this negative finding, the application of a novel data-driven technique for the analysis of the brain's global functional connectivity yielded a positive result. The weighted global connectivity (WGC) was significantly greater in the MCS group compared to the VS group (p < 0.05, family-wise error corrected). Using areas of significant WGC differences as 'seed regions' in a secondary connectivity analysis revealed extended functional networks in both MCS and healthy groups, whereas no such long-range functional connections were observed in the VS group. These results demonstrate the potential of functional connectivity MRI (fcMRI) as a clinical tool for differential diagnosis in disorders of consciousness.

Quantification of venous vessel size in human brain in response to hypercapnia and hyperoxia using magnetic resonance imaging.

Hypercapnia and hyperoxia give rise to vasodilation and vasoconstriction, respectively. This study investigates the influence of hypercapnia and hyperoxia on venous vessel size in the human brain. Venous vessel radii were measured in response to hypercapnia and hyperoxia. The venous vessel radii were determined by calculation of the changes in R2 * and R2 that are induced by breathing 6% CO2 or pure oxygen. The experimental paradigm consisted of two 3-min intervals of inhaling 6% CO2 or 100% O2 interleaved with three 2-min intervals of breathing air. Hypercapnic and hyperoxic experiments were performed on eight subjects on a 3T scanner. Parametric maps of mean venous vessel radius were calculated from the changes in R2 * and R2 , which were measured by simultaneous acquisition of gradient-echo and spin-echo signals. The mean venous vessel radii in hypercapnia were 7.3±0.3 µm in gray matter and 6.6±0.5 µm in white matter. The corresponding vessel radii in hyperoxia were 5.6±0.2 µm in gray matter and 5.4±0.2 µm in white matter. These results show that the venous vessel radius was larger in hypercapnia than that in hyperoxia in both gray matter and white matter (P<0.005), which agrees with the hypothesis that hypercapnia causes vasodilation and hyperoxia induces vasoconstriction.

A dual echo approach to motion correction for functional connectivity studies.

The effect of subject head movement on functional connectivity as measured by BOLD (blood oxygen level dependent) fMRI was investigated; movement mainly introduced increases in connectivity into the dataset. The effect of movement on connectivity is an important consideration when comparing patients suffering from neurological conditions to healthy controls, since it is well known that patients affected by such conditions are prone to move more in the scanner than healthy subjects. A method of motion correction utilising a dual echo EPI sequence is described. The first echo is acquired soon after the slice excitation (TE(1)=10 ms) when BOLD contrast is low and the MR signal is mainly sensitive to movement related effects, while the second echo is acquired at an echo time (TE(2)=30 ms) at which the MR signal is sensitive to both BOLD and movement related effects. To correct for additional signal variance introduced by subject movement, the second echo image is divided by the first echo image at each time point across the length of the scan. This procedure is easy to implement and requires no extra scan time. This method proved superior to the standard means of correction whereby realignment parameters and their first order derivatives are used as covariates of no interest in a linear regression model.

Electroconvulsive therapy reduces frontal cortical connectivity in severe depressive disorder.

To date, electroconvulsive therapy (ECT) is the most potent treatment in severe depression. Although ECT has been successfully applied in clinical practice for over 70 years, the underlying mechanisms of action remain unclear. We used functional MRI and a unique data-driven analysis approach to examine functional connectivity in the brain before and after ECT treatment. Our results show that ECT has lasting effects on the functional architecture of the brain. A comparison of pre- and posttreatment functional connectivity data in a group of nine patients revealed a significant cluster of voxels in and around the left dorsolateral prefrontal cortical region (Brodmann areas 44, 45, and 46), where the average global functional connectivity was considerably decreased after ECT treatment (P < 0.05, family-wise error-corrected). This decrease in functional connectivity was accompanied by a significant improvement (P < 0.001) in depressive symptoms; the patients' mean scores on the Montgomery Asberg Depression Rating Scale pre- and posttreatment were 36.4 (SD = 4.9) and 10.7 (SD = 9.6), respectively. The findings reported here add weight to the emerging "hyperconnectivity hypothesis" in depression and support the proposal that increased connectivity may constitute both a biomarker for mood disorder and a potential therapeutic target.

Vascular component analysis of hyperoxic and hypercapnic BOLD contrast.

Hyperoxia or hypercapnia provides a useful experimental tool to systematically alter the blood oxygenation level dependent (BOLD) contrast. Typical applications include calibrated functional magnetic resonance imaging (fMRI), BOLD sensitivity mapping, vessel size imaging or cerebrovascular reactivity mapping. This article describes a novel biophysical model of hyperoxic and hypercapnic BOLD contrast, which accounts for the magnetic susceptibility effects of molecular oxygen that is dissolved in blood and tissue, in addition to the well-established effects caused by the paramagnetic properties of deoxyhaemoglobin. Furthermore, the concept of vascular component analysis (VCA) is introduced and is shown to provide a computationally efficient tool for investigating the vascular specificity of hyperoxic and hypercapnic BOLD contrast. A theoretical investigation of gradient and spin echo BOLD contrast based on computer simulations was performed to compare three different conditions (hypercapnia induced by breathing 6% CO2, hyperoxia induced by breathing 100% O2, and simultaneous hypercapnia and hyperoxia induced by breathing carbogen, i.e. 5% CO2 in 95% CO2) with baseline (breathing air). Simulations were carried out for different levels of metabolic oxygen extraction fraction (OEF) ranging from 0 to 0.5. The key findings can be summarised as follows: (i) for hyperoxia the susceptibility of dissolved O2 may lead to a significant arterial BOLD contrast; (ii) under normoxic conditions the susceptibility of dissolved O2 is negligible; (iii) an almost complete loss of BOLD sensitivity may occur at lower OEF values in all parts of the vascular tree, whereas hyperoxic BOLD sensitivity is largely maintained; (iv) under hyperoxic conditions, a transition from positive to negative BOLD contrast occurs with decreasing OEF values. These findings have important implications for experimental applications of hyperoxic and hypercapnic BOLD contrast and may enable new clinical applications in ischemic stroke and other forms of acquired brain injury.

Magnetic resonance imaging of the mean venous vessel size in the human brain using transient hyperoxia.

Vessel size imaging is an emerging magnetic resonance imaging (MRI) technique which has been demonstrated to provide clinically relevant information about microvascular morphology. While previous studies of vessel size in humans relied on MRI contrast agents or hypercapnia-induced changes in blood oxygenation, the technique described here uses transient hyperoxia to alter the venous blood oxygenation. The experimental paradigm consisted of two 3-minute intervals of breathing 100% O(2) interleaved with three 2-minute intervals of breathing room air. Parametric maps of the mean venous vessel radius were calculated from changes in the blood oxygenation level dependent (BOLD) contrast which were measured using a combined spin-echo (SE) and gradient echo (GE) echo-planar imaging (EPI) sequence. The corresponding mean values in grey and white matter were r=6.5±0.3 µm and r=6.2±0.3 µm (n=6). While the hypercapnia technique requires a specialised gas mixture containing a low concentration of CO(2) (typically 5-6%), the hyperoxia technique presented here uses the inhalation of medical oxygen (100% O(2)) which is routinely available in a clinical environment. Furthermore, 100% O(2) is generally better tolerated than low doses of CO(2) which makes this technique particularly suitable for applications in critically ill patients.

Investigating brain response to music: a comparison of different fMRI acquisition schemes.

Functional magnetic resonance imaging (fMRI) in auditory experiments is a challenge, because the scanning procedure produces considerable noise that can interfere with the auditory paradigm. The noise might either mask the auditory material presented, or interfere with stimuli designed to evoke emotions because it sounds loud and rather unpleasant. Therefore, scanning paradigms that allow interleaved auditory stimulation and image acquisition appear to be advantageous. The sparse temporal sampling (STS) technique uses a very long repetition time in order to achieve a stimulus presentation in the absence of scanner noise. Although only relatively few volumes are acquired for the resulting data sets, there have been recent studies where this method has furthered remarkable results. A new development is the interleaved silent steady state (ISSS) technique. Compared with STS, this method is capable of acquiring several volumes in the time frame between the auditory trials (while the magnetization is kept in a steady state during stimulus presentation). In order to draw conclusions about the optimum fMRI procedure with auditory stimulation, different echo-planar imaging (EPI) acquisition schemes were compared: Continuous scanning, STS, and ISSS. The total acquisition time of each sequence was adjusted to about 12.5 min. The results indicate that the ISSS approach exhibits the highest sensitivity in detecting subtle activity in sub-cortical brain regions.

Evaluating an acoustically quiet EPI sequence for use in fMRI studies of speech and auditory processing.

Echoplanar MRI is associated with significant acoustic noise, which can interfere with the presentation of auditory stimuli, create a more challenging listening environment, and increase discomfort felt by participants. Here we investigate a scanning sequence that significantly reduces the amplitude of acoustic noise associated with echoplanar imaging (EPI). This is accomplished using a constant phase encoding gradient and a sinusoidal readout echo train to produce a narrow-band acoustic frequency spectrum, which is adapted to the scanner's frequency response function by choosing an optimum gradient switching frequency. To evaluate the effect of these nonstandard parameters we conducted a speech experiment comparing four different EPI sequences: Quiet, Sparse, Standard, and Matched Standard (using the same readout duration as Quiet). For each sequence participants listened to sentences and signal-correlated noise (SCN), which provides an unintelligible amplitude-matched control condition. We used BOLD sensitivity maps to quantify sensitivity loss caused by the longer EPI readout duration used in the Quiet and Matched Standard EPI sequences. We found that the Quiet sequence provided more robust activation for SCN in primary auditory areas and comparable activation in frontal and temporal regions for Sentences>SCN, but less sentence-related activity in inferotemporal cortex. The increased listening effort associated with the louder Standard sequence relative to the Quiet sequence resulted in increased activation in the left temporal and inferior parietal cortices. Together, these results suggest that the Quiet sequence is suitable, and perhaps preferable, for many auditory studies. However, its applicability depends on the specific brain regions of interest.

Dual echo EPI--the method of choice for fMRI in the presence of magnetic field inhomogeneities?

FMRI studies of the orbitofrontal cortex or the inferior temporal lobes are often compromised by susceptibility artefacts, which may result in signal reduction or loss in gradient echo (GE) EPI. Spin echo (SE) EPI is considerably more robust against susceptibility-related signal loss, but its intrinsic sensitivity to changes in the blood oxygenation level dependent (BOLD) contrast is generally lower. In this study, we performed a direct comparison of GE and SE fMRI using a single-shot dual echo EPI acquisition scheme. Transient hypercapnia, induced by breathing Carbogen (5% CO(2), 95% O(2)), was used as a global physiological stimulus to alter the BOLD contrast. In regions affected by magnetic field inhomogeneities, SE EPI provided significantly higher BOLD sensitivity than GE EPI. Such regions included the orbitofrontal cortex, temporal pole, anterior inferior temporal cortex, as well as parts of the lateral inferior temporal cortex and the lateral cerebellum. Dual echo fMRI benefits from the robustness of SE EPI in these critical regions while utilising the generally higher sensitivity of GE EPI in normal regions. It therefore provides an attractive solution for fMRI studies that require optimum sensitivity in both normal and critical brain regions. Furthermore, a general method is proposed to combine the GE and SE data into a single hybrid data set that provides optimum sensitivity in the whole brain. This method can be applied to any experimental design that can be expressed in terms of a generalised linear model. The feasibility of this approach is demonstrated both theoretically and experimentally.