Multicenter Imaging Research Articles

Investigating the state-of-the-art in whole-body MR-based attenuation correction: an intra-individual, inter-system, inventory study on three clinical PET/MR systems.

We assess inter- and intra-subject variability of magnetic resonance (MR)-based attenuation maps (MRμMaps) of human subjects for state-of-the-art positron emission tomography (PET)/MR imaging systems. Four healthy male subjects underwent repeated MR imaging with a Siemens Biograph mMR, Philips Ingenuity TF and GE SIGNA PET/MR system using product-specific MR sequences and image processing algorithms for generating MRμMaps. Total lung volumes and mean attenuation values in nine thoracic reference regions were calculated. Linear regression was used for comparing lung volumes on MRμMaps. Intra- and inter-system variability was investigated using a mixed effects model. Intra-system variability was seen for the lung volume of some subjects, (p = 0.29). Mean attenuation values across subjects were significantly different (p < 0.001) due to different segmentations of the trachea. Differences in the attenuation values caused noticeable intra-individual and inter-system differences that translated into a subsequent bias of the corrected PET activity values, as verified by independent simulations. Significant differences of MRμMaps generated for the same subjects but different PET/MR systems resulted in differences in attenuation correction factors, particularly in the thorax. These differences currently limit the quantitative use of PET/MR in multi-center imaging studies.

Heritability and reliability of automatically segmented human hippocampal formation subregions

The human hippocampal formation can be divided into a set of cytoarchitecturally and functionally distinct subregions, involved in different aspects of memory formation. Neuroanatomical disruptions within these subregions are associated with several debilitating brain disorders including Alzheimer's disease, major depression, schizophrenia, and bipolar disorder. Multi-center brain imaging consortia, such as the Enhancing Neuro Imaging Genetics through Meta-Analysis (ENIGMA) consortium, are interested in studying disease effects on these subregions, and in the genetic factors that affect them. For large-scale studies, automated extraction and subsequent genomic association studies of these hippocampal subregion measures may provide additional insight. Here, we evaluated the test–retest reliability and transplatform reliability (1.5T versus 3T) of the subregion segmentation module in the FreeSurfer software package using three independent cohorts of healthy adults, one young (Queensland Twins Imaging Study, N=39), another elderly (Alzheimer's Disease Neuroimaging Initiative, ADNI-2, N=163) and another mixed cohort of healthy and depressed participants (Max Planck Institute, MPIP, N=598). We also investigated agreement between the most recent version of this algorithm (v6.0) and an older version (v5.3), again using the ADNI-2 and MPIP cohorts in addition to a sample from the Netherlands Study for Depression and Anxiety (NESDA) (N=221). Finally, we estimated the heritability (h2) of the segmented subregion volumes using the full sample of young, healthy QTIM twins (N=728). Test–retest reliability was high for all twelve subregions in the 3T ADNI-2 sample (intraclass correlation coefficient (ICC)=0.70–0.97) and moderate-to-high in the 4T QTIM sample (ICC=0.5–0.89). Transplatform reliability was strong for eleven of the twelve subregions (ICC=0.66–0.96); however, the hippocampal fissure was not consistently reconstructed across 1.5T and 3T field strengths (ICC=0.47–0.57). Between-version agreement was moderate for the hippocampal tail, subiculum and presubiculum (ICC=0.78–0.84; Dice Similarity Coefficient (DSC)=0.55–0.70), and poor for all other subregions (ICC=0.34–0.81; DSC=0.28–0.51). All hippocampal subregion volumes were highly heritable (h2=0.67–0.91). Our findings indicate that eleven of the twelve human hippocampal subregions segmented using FreeSurfer version 6.0 may serve as reliable and informative quantitative phenotypes for future multi-site imaging genetics initiatives such as those of the ENIGMA consortium.

Construction of brain atlases based on a multi-center MRI dataset of 2020 Chinese adults.

Despite the known morphological differences (e.g., brain shape and size) in the brains of populations of different origins (e.g., age and race), the Chinese brain atlas is less studied. In the current study, we developed a statistical brain atlas based on a multi-center high quality magnetic resonance imaging (MRI) dataset of 2020 Chinese adults (18–76 years old). We constructed 12 Chinese brain atlas from the age 20 year to the age 75 at a 5 years interval. New Chinese brain standard space, coordinates, and brain area labels were further defined. The new Chinese brain atlas was validated in brain registration and segmentation. It was found that, as contrast to the MNI152 template, the proposed Chinese atlas showed higher accuracy in hippocampus segmentation and relatively smaller shape deformations during registration. These results indicate that a population-specific time varying brain atlas may be more appropriate for studies involving Chinese populations.

Multi-Atlas Library for Eliminating Normalization Failures in Non-Human Primates.

Current tools for automated skull stripping, normalization, and segmentation of non-human primate (NHP) brain MRI studies typically demonstrate high failure rates. Many of these failures are due to a poor initial estimate for the affine component of the transformation. The purpose of this study is to introduce a multi-atlas approach to overcome these limitations and drive the failure rate to near zero. A library of study-specific templates (SST) spanning three Old World primate species (Macaca fascicularis, M. mulatta, Chlorocebus aethiops) was created using a previously described unbiased automated approach. Several modifications were introduced to the methodology to improve initial affine estimation at the study-specific template level, and at the individual subject level. These involve performing multiple separate normalizations to a multi-atlas library of templates and selecting the best performing template on the basis of a covariance similarity metric. This template was then used as an initialization for the affine component of subsequent skull stripping and normalization procedures. Normalization failure rate for SST generation and individual-subject segmentation on a set of 150 NHP was evaluated on the basis of visual inspection. The previous automated template creation procedure results in excellent skull stripping, segmentation, and atlas labeling across species. Failure rate at the individual-subject level was approximately 1%, however at the SST generation level it was 17%. Using the new multi-atlas approach, failure rate was further reduced to zero for both SST generation and individual subject processing. We describe a multi-atlas library registration approach for driving normalization failures in NHP to zero. It is straightforward to implement, and can have application to a wide variety of existing tools, as well as in difficult populations including neonates and the elderly. This approach is also an important step towards developing fully automated high-throughput processing pipelines that are critical for future high volume multi-center NHP imaging studies for studies of drug abuse and brain health.

Abstract 14051: Changes in Medical Therapy and Lifestyle After Anatomical Versus Functional Testing for Coronary Artery Disease: The PROMISE Trial (PROspective Multicenter Imaging Study for Evaluation of Chest Pain)

Introduction: Diagnostic testing may represent a “teachable moment” for patients newly presenting with symptoms suggestive of CAD and requiring further evaluation, and may influence risk factor management, independent of test results. However, little is known about changes in medications and lifestyle after anatomical versus functional testing. Hypothesis: Patients assigned to coronary CTA will report greater use of preventive medications and lifestyle practices. Methods: We randomly assigned 10,003 symptomatic patients (53% women; mean age 61 yrs) to anatomical testing with CTA or functional testing (exercise ECG, nuclear stress, or stress echocardiography). We assessed use of preventive medications (aspirin, statin, beta blocker, and ACEi/ARB) and lifestyle practices (heart healthy diet, regular exercise, smoking, and obese/overweight status [BMI&gt;25]) at 60 days. Chi-square tests assessed between-group changes (initiation or discontinuation). Multivariable logistic regression models assessed the association between testing strategy and prevalence of medication use or lifestyle practices. Results: There were no differences in medications or lifestyle practices at baseline. At 60 days, the CTA strategy was associated with a higher proportion of patients newly initiating aspirin (11.6% vs 7.6%), statin (12.7% vs 6.2%), and beta blockers (8.2% vs 5.4%), compared to functional testing (p&lt;0.0001 for each). The CTA strategy was also associated with a higher incidence of weight loss among obese/overweight patients (2.8% vs 2.4%), but the difference was not significant (p=0.051). Overall prevalences of aspirin (aOR=1.55), statin (aOR=2.04), and beta blocker (aOR=1.32) use were higher after CTA (p&lt;0.001 for each). Eating a healthy diet (54.7% vs 50.8%) was increased after CTA (aOR=1.13 p=0.004), whereas obese/overweight status was reduced (aOR=0.79 p=0.047). Exercise and smoking cessation increased similarly in both arms. Benefits of CTA for statin use and weight loss persisted after adjusting for test results. Conclusions: Among patients with suspected CAD, anatomical testing is associated with greater favorable changes in preventive medical and lifestyle practices. This may represent a long term benefit of a CTA testing strategy.

Abstract 10369: Safety of Anatomic versus Functional Diagnostic Testing: Insights From the PROMISE Trial (PROspective Multicenter Imaging Study for Evaluation of Chest Pain)

Introduction: Suspected coronary artery disease (CAD) is a common indication for diagnostic testing. However, a comprehensive, randomized comparison of test safety, including radiation dose, incidental findings, and test complications, of coronary computed tomography angiography (CTA) and functional tests has not been performed. Methods: All PROMISE patients completing a diagnostic test were analyzed as tested for differences between CTA vs functional tests in A) 90 day cumulative radiation dose in milliSieverts (mSv), B) index test and cumulative dose for those whose physicians intended referral for nuclear stress before randomization, C) incidental findings (all and prespecified as potentially clinically significant) and D) index test complications. Results: Of 9470 eligible patients (mean age 61, 53% women), 4633 had CTA and 4837 had functional testing (3263 nuclear stress, 1083 stress echo and 491 exercise ECG). As a strategy comparison, CTA had lower median but higher mean cumulative radiation dose than functional testing (median: 10.3 vs 11.5 mSv, p&lt;0.001, mean: 12.5 vs 10.6, p&lt;0.001). In a test to test comparison in 6349 intended for nuclear stress before randomization, CTA had lower index test (mean: 10.5 vs 14.1, p&lt;0.001) and cumulative (mean: 12.4 vs 15.2, p&lt;0.001) dose than nuclear stress. Incidental findings were more common on CTA than functional tests (43% vs 2%, p&lt;0.001). CTA found more potentially clinically significant incidentals (19% vs 0.7%, p&lt;0.001), including valvular disease (echo), breast uptake (nuclear), and lung nodules, pneumonia or pulmonary embolism (CT). There were no major index test complications. Minor complications occurred in 0.8% of CTAs and 0.4% of functional tests (p&lt;0.01). In those intended for nuclear stress, complications occurred in 0.8% of CTAs and 0.5% of nuclear stress tests (p=0.17), including hypotension (nuclear) and mild contrast reaction or extravasation (CT). No CTA and 5 functional test complications led to hospitalization. Conclusions: CTA had lower median but higher mean radiation dose compared to all functional testing, but lower dose than nuclear stress. CTA had more potentially clinically meaningful incidental findings, and despite very low overall rates, more minor complications.

Abstract 18053: Prognostic Value of Anatomic versus Functional Diagnostic Testing in Symptomatic Patients With Suspected CAD: The PROMISE Trial (PROspective Multicenter Imaging Study for Evaluation of Chest Pain)

Introduction: Symptoms suggestive of coronary artery disease (CAD) are common, but no randomized comparison of the prognostic value of coronary computed tomographic angiography (CTA) vs. functional testing has been performed. Methods: Patients (10,003) with suspected CAD were randomized to functional testing (exercise ECG, nuclear stress, or stress echo) or CTA. Site-based diagnostic test reports were classified as normal or mildly, moderately, or severely abnormal. Cox proportional hazards models assessed the relationship of test results to a composite of time to death, myocardial infarction, or unstable angina, and the discriminatory ability of the two tests was compared using the C-index. Adjusted HRs for events in patients with abnormal vs. normal test results were determined for CTA and functional testing. Results: Overall, 4516 patients received CTA and 4572 functional testing as the randomized diagnostic test. The proportions of normal, mild, moderate, and severe abnormalities on CTA were 33.3%, 42.7%, 10.0%, and 14.0%, respectively, and for functional testing 78.9%, 8.5%, 4.2%, and 8.4% (P=0.001). Event rates over 26.0 months’ (IQR 18.0, 34.4) follow-up were similar: CTA 137/4516 (3.0%) vs. functional 132/4572 (2.9%). Compared to patients with normal CTAs, those with abnormal CTAs had increasing HRs for events with increasingly severe abnormalities, while only patients with moderate and severely abnormal functional testing had an increased HR for events (Figure). The probability of having an event given a normal test result was lower for CTA compared to functional testing (0.9% vs. 2.2%; HR 0.47 [95% CI 0.26, 0.83]; P=0.009), but not significantly different for other test strata. The overall prognostic performance (C-index) of CTA vs. functional testing was 0.72 (0.68, 0.76) vs. 0.64 (0.60, 0.69); P=0.022 (see KM curves in Figure). Conclusions: In symptomatic patients with suspected CAD, CTA has better prognostic value than functional testing.

Morphological Abnormalities of Thalamic Subnuclei in Migraine: A Multicenter MRI Study at 3 Tesla.

This multicenter imaging study shows morphological thalamic abnormalities in a large cohort of patients with episodic migraine compared with healthy subjects using state-of-the-art MRI and advanced, fully automated multiatlas segmentation techniques. The results stress that migraine is a disorder of the CNS in which not only is brain function abnormal, but also brain structure is undergoing significant remodeling.

TCT-284 Comparison of Visual Assessment of Coronary Stenosis with Independent Quantitative Coronary Angiography: Findings from the PROMISE Trial (PROspective Multicenter Imaging Study for Evaluation of chest pain)

TCT-284 Comparison of Visual Assessment of Coronary Stenosis with Independent Quantitative Coronary Angiography: Findings from the PROMISE Trial (PROspective Multicenter Imaging Study for Evaluation of chest pain)

Age at First Episode Modulates Diagnosis-Related Structural Brain Abnormalities in Psychosis.

Brain volume and thickness abnormalities have been reported in first-episode psychosis (FEP). However, it is unclear if and how they are modulated by brain developmental stage (and, therefore, by age at FEP as a proxy). This is a multicenter cross-sectional case-control brain magnetic resonance imaging (MRI) study. Patients with FEP (n = 196), 65.3% males, with a wide age at FEP span (12-35 y), and healthy controls (HC) (n = 157), matched for age, sex, and handedness, were scanned at 6 sites. Gray matter volume and thickness measurements were generated for several brain regions using FreeSurfer software. The nonlinear relationship between age at scan (a proxy for age at FEP in patients) and volume and thickness measurements was explored in patients with schizophrenia spectrum disorders (SSD), affective psychoses (AFP), and HC. Earlier SSD cases (ie, FEP before 15-20 y) showed significant volume and thickness deficits in frontal lobe, volume deficits in temporal lobe, and volume enlargements in ventricular system and basal ganglia. First-episode AFP patients had smaller cingulate cortex volume and thicker temporal cortex only at early age at FEP (before 18-20 y). The AFP group also had age-constant (12-35-y age span) volume enlargements in the frontal and parietal lobe. Our study suggests that age at first episode modulates the structural brain abnormalities found in FEP patients in a nonlinear and diagnosis-dependent manner. Future MRI studies should take these results into account when interpreting samples with different ages at onset and diagnosis.

Multicenter imaging outcomes study of The Cancer Genome Atlas glioblastoma patient cohort: imaging predictors of overall and progression-free survival.

Despite an aggressive therapeutic approach, the prognosis for most patients with glioblastoma (GBM) remains poor. The aim of this study was to determine the significance of preoperative MRI variables, both quantitative and qualitative, with regard to overall and progression-free survival in GBM. We retrospectively identified 94 untreated GBM patients from the Cancer Imaging Archive who had pretreatment MRI and corresponding patient outcomes and clinical information in The Cancer Genome Atlas. Qualitative imaging assessments were based on the Visually Accessible Rembrandt Images feature-set criteria. Volumetric parameters were obtained of the specific tumor components: contrast enhancement, necrosis, and edema/invasion. Cox regression was used to assess prognostic and survival significance of each image. Univariable Cox regression analysis demonstrated 10 imaging features and 2 clinical variables to be significantly associated with overall survival. Multivariable Cox regression analysis showed that tumor-enhancing volume (P = .03) and eloquent brain involvement (P < .001) were independent prognostic indicators of overall survival. In the multivariable Cox analysis of the volumetric features, the edema/invasion volume of more than 85 000 mm(3) and the proportion of enhancing tumor were significantly correlated with higher mortality (Ps = .004 and .003, respectively). Preoperative MRI parameters have a significant prognostic role in predicting survival in patients with GBM, thus making them useful for patient stratification and endpoint biomarkers in clinical trials.

Quality Assurance and Control Issues in Multi-Center Diffusion Tensor Imaging Studies

Quality Assurance and Control Issues in Multi-Center Diffusion Tensor Imaging Studies

The PROMISE study: a clear promise for functional stress testing in patients with suspected coronary artery disease.

Over the years, considerable debate has arisen whether a pure anatomical test would suffice to demonstrate the significance of coronary artery disease (CAD) in patients suspected for CAD, and–more importantly–whether this anatomical information would have a bearing on clinical outcome [1, 2]. In particular, coronary computed tomography angiography (CTA) has been put forward as an optimal non-invasive anatomical imaging test to detect CAD in patients with stable CAD [3]. A number of studies has shown that CTA has a high sensitivity, reasonable specificity and an extremely high negative predictive value [4–10]. Several large-scale studies have shown that a strategy of CTA use in the emergency department is associated with faster discharge, as compared with standard care, without a significant difference in event rates [11–13]. However, the lack of evidence supporting CTA in randomised trials has also been mentioned. As a consequence, the relative impact of data from non-invasive anatomical testing versus functional testing on subsequent management and clinical outcomes is not fully known. At the recently held conference of the American College of Cardiology (ACC), San Diego, California, 14–18 March 2015, the Prospective Multicenter Imaging Study for Evaluation of Chest Pain (PROMISE) trial was presented by Pamela S. Douglas, MD (Duke University School of Medicine, Durham, North Carolina, USA) and simultaneously published online in The New England Journal of Medicine [14]. The goal of the PROMISE trial was to evaluate anatomical testing using CTA compared with functional testing among low- to intermediate-risk patients with chest pain suspicious for CAD. The primary hypothesis of the study was that the clinical outcomes in patients assigned to anatomical testing with the use of CTA would be superior to those in patients assigned to functional testing. A total of 10,003 low- to intermediate-risk patients with chest pain (mean age 61 years, 53 % female patients, 21 % diabetics) were randomised to evaluation with an anatomical strategy (n = 4996) versus a functional strategy (n = 5007). Patients randomised to an anatomical strategy underwent a 64-slice CTA, while patients randomised to a functional strategy underwent exercise electrocardiography (ECG), exercise imaging or pharmacological stress imaging. Among either group who underwent a functional test, 68 % underwent nuclear stress testing, 22 % underwent stress echocardiography and 10 % underwent exercise ECG. Duration of follow-up was a median of 25 months. The primary outcome, all-cause mortality, myocardial infarction, hospitalisation for unstable angina, or major complication from a cardiovascular procedure occurred in 3.3 % of the anatomical testing group versus 3.0 % of the functional testing group (p = 0.75). Among low- to intermediate-risk patients with chest pain, anatomical testing with coronary CTA was not superior to functional testing. CTA was associated with an increased frequency of cardiac catheterisation; however, it was associated with a lower frequency of invasive catheterisation showing non-obstructive CAD. Anatomical testing was also associated with increased radiation exposure and a non-significant increase in total costs. In conclusion, in symptomatic patients with suspected CAD who required non-invasive testing, an initial strategy of CTA was not associated with better clinical outcomes than functional testing over a median follow-up of 2 years. Simply stated, the PROMISE study suggests therefore that patients, suspected for CAD and undergoing CTA, do not have less risk of heart attack, dying or being hospitalised months later than those who take a simple treadmill test or other functional test. According to W. Douglas Weaver, MD, former president of the ACC, these findings should temper the enthusiastic use of CTA to screen patients with chest pain–it is not worth the added radiation and use of unnecessary heart catheterisations and stent implantations, which did nothing to improve the outcome of patients. Valentin Fuster, MD, current editor-in-chief of the Journal of the American College of Cardiology questioned if there could be a long-term benefit in direct visualisation using CTA. The study author, Pamela S. Douglas, said the research group plans to further investigate outcomes for different subgroups of patients to determine whether different groups might benefit from different testing approaches. In an accompanying Editorial [15], it was stated that the International Study of Comparative Health Effectiveness with Medical and Invasive Approaches (ISCHEMIA; ClinicalTrials.gov number, NCT01471522), in which randomised therapy (invasive versus medical, which is driven by the presence of extensive ischaemia on functional stress testing), will help answer this question. So far, the PROMISE trial offers clear promise that functional stress testing provides at least similar information to CTA and–for economical and safety reasons–might prevail over CTA alone in patients with suspected CAD.

Noninvasive testing strategies in symptomatic, intermediate-risk CAD patients: a perspective on the "PROMISE" trial and its potential implementation in clinical practice.

While the results of the Prospective Multicenter Imaging Study for Evaluation of Chest Pain trial (PROMISE trial) are negative for the primary outcome, the results from this large, contemporary trial of >10,000 patients provide important insights into clinical management of patients presenting with chest pain. The results reinforce that while diagnostic testing is an important component of modern management, its choice should be directed by a clinician in a clinical context and with subsequent management in mind. Based on presentation and pre-test probability, the clinician will decide if any additional testing necessary is necessary and if that is the case chose the most appropriate test according to current guidelines, applied to the individual patient and clinical scenario.

Periventricular hyperintensity in children with hydrocephalus.

Magnetic resonance images of children with hydrocephalus often include a rim of hyperintensity in the periventricular white matter (halo). The purpose of this study was to decide between the hypothesis that the halo is caused by cerebrospinal fluid (CSF) flow during the cardiac cycle, and the alternate hypothesis that the halo is caused by anatomical changes (stretching and compression of white matter). Participants were selected from a multicenter imaging study of pediatric hydrocephalus. We compared 19 children with hydrocephalus to a group of 52 controls. We quantified ventricle enlargement using the frontal-occipital horn ratio. We conducted qualitative and quantitative analysis of diffusion tensor imaging in the corpus callosum and posterior limb of the internal capsule. Parameters included the fractional anisotropy (FA), mean diffusivity, axial diffusivity and radial diffusivity. The halo was seen in 16 of the 19 children with hydrocephalus but not in the controls. The corpus callosum of the hydrocephalus group demonstrated FA values that were significantly decreased from those in the control group (P = 4 · 10(-6)), and highly significant increases were seen in the mean diffusivity and radial diffusivity in the hydrocephalus group. In the posterior limb of the internal capsule the FA values of the hydrocephalus group were higher than those for the control group (P = 0.002), and higher values in the hydrocephalus group were also noted in the axial diffusivity. We noted correlations between the diffusion parameters and the frontal-occipital horn ratio. Our results strongly support the hypothesis that the halo finding in hydrocephalus is caused by structural changes rather than pulsatile CSF flow.

Abstract T P169: Acute Cardioembolic vs. Large Artery Atherosclerotic Infarction-related Lesion Topography and Right-Left Propensity: A Multi-center Quantitative Magnetic Resonance Imaging Study

Background and Purpose: A recent study showed that cardiogenic emboli might flow more frequently into the right hemisphere, whereas atheromatous aortic arch emboli might flow more frequently into the left hemisphere. We tried to 1) see if cardioembolic (CE) infarct volume would be larger in the right hemisphere than in the left hemisphere, and 2) depict anatomical regions showing CE vs. large artery atherosclerotic (LAA) infarction-related right-left propensity. Methods: In this study on carotid artery territory CE (n = 694) vs. LAA (n = 1162) acute ischemic stroke patients who were enrolled consecutively from 11 nationwide stroke centers, we quantitatively registered diffusion magnetic resonance imaging lesions onto the Montreal Neurologic Institute brain template. Results: In patients with bilateral CE stroke (n = 163), right hemispheric infarct size was about two times bigger than the contralateral left hemispheric infarct size (p = 0.002). However, in patients with either unilateral (n = 925) or bilateral (n = 184) LAA stroke, there was no significant difference in the right vs. left hemispheric infarct size between the right vs. left unilateral infarct groups or within the bilateral infarct group (all p &gt; 0.05). In patients with unilateral CE stroke (n = 510), there was no significant difference in the infarct size between the groups with right vs. left hemispheric lesions. Age and infarct volume-adjusted p-value maps of the CE vs. LAA stroke patients, which were corrected for multiple comparisons, revealed the brain regions with a significantly higher infarct frequency in CE stroke than in LAA stroke. The significant clusters were observed only in the right hemisphere, encompassing both the superior and inferior division middle cerebral artery territories, particularly in the pial and corticosubcortical regions including the claustrum and insula. Conclusion: The present multi-center quantitative magnetic resonance imaging study confirms the ‘right &gt; left’ propensity of CE (vs. LAA) stroke.

Stable Atlas-based Mapped Prior (STAMP) machine-learning segmentation for multicenter large-scale MRI data

Machine learning (ML)-based segmentation methods are a common technique in the medical image processing field. In spite of numerous research groups that have investigated ML-based segmentation frameworks, there remains unanswered aspects of performance variability for the choice of two key components: ML algorithm and intensity normalization. This investigation reveals that the choice of those elements plays a major part in determining segmentation accuracy and generalizability. The approach we have used in this study aims to evaluate relative benefits of the two elements within a subcortical MRI segmentation framework. Experiments were conducted to contrast eight machine-learning algorithm configurations and 11 normalization strategies for our brain MR segmentation framework. For the intensity normalization, a Stable Atlas-based Mapped Prior (STAMP) was utilized to take better account of contrast along boundaries of structures. Comparing eight machine learning algorithms on down-sampled segmentation MR data, it was obvious that a significant improvement was obtained using ensemble-based ML algorithms (i.e., random forest) or ANN algorithms. Further investigation between these two algorithms also revealed that the random forest results provided exceptionally good agreement with manual delineations by experts. Additional experiments showed that the effect of STAMP-based intensity normalization also improved the robustness of segmentation for multicenter data sets. The constructed framework obtained good multicenter reliability and was successfully applied on a large multicenter MR data set (n>3000). Less than 10% of automated segmentations were recommended for minimal expert intervention. These results demonstrate the feasibility of using the ML-based segmentation tools for processing large amount of multicenter MR images. We demonstrated dramatically different result profiles in segmentation accuracy according to the choice of ML algorithm and intensity normalization chosen.

Influence of Stroke Infarct Location on Functional Outcome Measured by the Modified Rankin Scale

In the early days after ischemic stroke, information on structural brain damage from MRI supports prognosis of functional outcome. It is rated widely by the modified Rankin Scale that correlates only moderately with lesion volume. We therefore aimed to elucidate the influence of lesion location from early MRI (days 2-3) on functional outcome after 1 month using voxel-based lesion symptom mapping. We analyzed clinical and MRI data of patients from a prospective European multicenter stroke imaging study (I-KNOW). Lesions were delineated on fluid-attenuated inversion recovery images on days 2 to 3 after stroke onset. We generated statistic maps of lesion contribution related to clinical outcome (modified Rankin Scale) after 1 month using voxel-based lesion symptom mapping. Lesion maps of 101 patients with middle cerebral artery infarctions were included for analysis (right-sided stroke, 47%). Mean age was 67 years, median admission National Institutes of Health Stroke Scale was 11. Mean infarct volumes were comparable between both sides (left, 37.5 mL; right, 43.7 mL). Voxel-based lesion symptom mapping revealed areas with high influence on higher modified Rankin Scale in regions involving the corona radiata, internal capsule, and insula. In addition, asymmetrically distributed impact patterns were found involving the right inferior temporal gyrus and left superior temporal gyrus. In this group of patients with stroke, characteristic lesion patterns in areas of motor control and areas involved in lateralized brain functions on early MRI were found to influence functional outcome. Our data provide a novel map of the impact of lesion localization on functional stroke outcome as measured by the modified Rankin Scale.

Cost analysis favours SPECT over PET and CTA for evaluation of coronary artery disease: the SPARC study.

There are many ways to perform noninvasive testing in patients with coronary artery disease (CAD), including myocardial perfusion imaging with single-photon emission computed tomography (SPECT), positron-emission tomography (PET), and coronary computed tomography angiography (CTA). Results from series of patients evaluated with single modalities have already been published by many researchers [1–3], but there are few head-to-head comparisons of the outcomes of alternative test strategies [4, 5]. Consequently, the comparative effectiveness of different noninvasive cardiac testing strategies has been difficult to assess. The recently published SPARC registry (Study of Myocardial Perfusion and Coronary Anatomy Imaging Roles in Coronary Artery Disease) in JACC by the group of DiCarli et al. (Stanford, CA, USA) was a multicentre study designed to collect standardised clinical data on patients undergoing CTA, PET, or SPECT [6]. The primary purpose of the SPARC study was to evaluate the economic outcomes of using CTA, PET, or SPECT to evaluate patients with suspected CAD. The authors used an observational registry of 1703 patients: 590 patients underwent CTA, 548 patients had PET and 565 patients SPECT for diagnosis of suspected CAD. Patients were followed for 2 years; documented resource use, medical costs for CAD, and clinical outcomes were assessed. The main findings were that costs were significantly lower for SPECT compared with CTA or PET in the evaluation of suspected CAD. SPECT was economically attractive compared with PET, whereas CTA was associated with higher costs and no significant difference in mortality compared with SPECT. More specifically, the authors observed that patients who underwent coronary CTA and PET experienced higher rates of subsequent invasive coronary arteriography (16 % and 15 %, respectively) as compared with patients who underwent SPECT (7 %). The mean total radiation was significantly higher for CTA (15.1 mSV) compared with SPECT (11.7 mSV), both for the initial tests (13 mSV vs. 11 mSV) and for follow-up tests and procedures (2.1 mSV vs. 0.7 mSV). Exposure from follow-up tests and procedures was higher after PET (2.0 mSV) than after SPECT (0.6 mSV). The findings in favour of SPECT are in line with a decision model by Garber and Solomon [7], who compared exercise treadmill testing, stress echocardiography, planar thallium imaging, SPECT, and PET imaging with a strategy of immediate invasive angiography. They found SPECT to be much more cost-effective than PET for noninvasive diagnosis, and SPECT to be a better option than immediate invasive coronary angiography. In terms of outcomes, Shaw et al. [1] found that patients who underwent initial SPECT with selective cardiac catheterisation had lower costs than patients who underwent routine coronary angiography (END study). In the EMPIRE study by Underwood et al. [2] it was demonstrated that strategies using myocardial perfusion imaging were cheaper and equally effective when compared with strategies that did not use myocardial perfusion imaging, both for the cost of diagnosis and for overall 2-year management costs; the 2-year patient outcome was the same. Sharples et al. [8] randomised 898 patients to SPECT, stress echocardiography, magnetic resonance imaging (MRI), or direct invasive coronary angiography, and found SPECT to be as useful as immediate invasive angiography with similar costs. A limitation of the SPARC study is the fact that MRI was not included in the comparative strategies. In addition, as the study started in 2006, in recent years many technological revolutions have taken place for all imaging modalities indicating that studies such as SPARC should be repeated and extended every 5–10 years [9]. Analysis of the current SPARC registry at least suggests that SPECT is the noninvasive imaging procedure of choice to evaluate patients with suspected CAD in terms of subsequent medical costs. SPECT offered the lowest costs over 2 years of follow-up among patients with suspected CAD. The significantly higher costs among patients undergoing CTA or PET were primarily due to higher rates of subsequent invasive cardiac procedures, because there was little difference in initial costs of testing. To conclude, the use of SPECT myocardial perfusion imaging was associated with lower costs over 2 years of follow-up than the use of CTA or PET, primarily because of fewer subsequent invasive procedures. Consequently, SPECT imaging seems to be the preferred imaging strategy in patients with suspected CAD. The value of SPECT imaging has been underscored by several recent reports [10–13]. However, larger and more definitive studies, such as the ongoing PROMISE (Prospective Multicenter Imaging Study for Evaluation of Chest Pain) trial (NCT01174550), are needed to assess the value and cost effectiveness of the various noninvasive imaging strategies.

PROspective Multicenter Imaging Study for Evaluation of chest pain: Rationale and design of the PROMISE trial

Suspected coronary artery disease (CAD) is one of the most common, potentially life-threatening diagnostic problems clinicians encounter. However, no large outcome-based randomized trials have been performed to guide the selection of diagnostic strategies for these patients. The PROMISE study is a prospective, randomized trial comparing the effectiveness of 2 initial diagnostic strategies in patients with symptoms suspicious for CAD. Patients are randomized to either (1) functional testing (exercise electrocardiogram, stress nuclear imaging, or stress echocardiogram) or (2) anatomical testing with ≥64-slice multidetector coronary computed tomographic angiography. Tests are interpreted locally in real time by subspecialty certified physicians, and all subsequent care decisions are made by the clinical care team. Sites are provided results of central core laboratory quality and completeness assessment. All subjects are followed up for ≥1 year. The primary end point is the time to occurrence of the composite of death, myocardial infarction, major procedural complications (stroke, major bleeding, anaphylaxis, and renal failure), or hospitalization for unstable angina. More than 10,000 symptomatic subjects were randomized in 3.2 years at 193 US and Canadian cardiology, radiology, primary care, urgent care, and anesthesiology sites. Multispecialty community practice enrollment into a large pragmatic trial of diagnostic testing strategies is both feasible and efficient. The PROMISE trial will compare the clinical effectiveness of an initial strategy of functional testing against an initial strategy of anatomical testing in symptomatic patients with suspected CAD. Quality of life, resource use, cost-effectiveness, and radiation exposure will be assessed.