Assessing the Performance of Artificial Intelligence Models: Insights from the American Society of Functional Neuroradiology Artificial Intelligence Competition.

BACKGROUND AND PURPOSE: Artificial intelligence (AI) models in radiology are frequently developed and validated using datasets from a single institution and are rarely tested on independent, external datasets, raising questions about their generalizability and applicability in clinical practice. The American Society of Functional Neuroradiology (ASFNR) organized a multi-center AI competition to evaluate the proficiency of developed models in identifying various pathologies on NCCT, assessing age-based normality and estimating medical urgency. MATERIALS AND METHODS: In total, 1201 anonymized, full-head NCCT clinical scans from five institutions were pooled to form the dataset. The dataset encompassed normal studies as well as pathologies including acute ischemic stroke, intracranial hemorrhage, traumatic brain injury, and mass effect (detection of these-task 1). NCCTs were also assessed to determine if findings were consistent with expected brain changes for the patient's age (task 2: age-based normality assessment) and to identify any abnormalities requiring immediate medical attention (task 3: evaluation of findings for urgent intervention). Five neuroradiologists labeled each NCCT, with consensus interpretations serving as the ground truth. The competition was announced online, inviting academic institutions and companies. Independent central analysis assessed each model's performance. Accuracy, sensitivity, specificity, positive and negative predictive values, and receiver operating characteristic (ROC) curves were generated for each AI model, along with the area under the ROC curve (AUROC). RESULTS: 1177 studies were processed by four teams. The median age of patients was 62, with an interquartile range of 33. 19 teams from various academic institutions registered for the competition. Of these, four teams submitted their final results. No commercial entities participated in the competition. For task 1, AUROCs ranged from 0.49 to 0.59. For task 2, two teams completed the task with AUROC values of 0.57 and 0.52. For task 3, teams had little to no agreement with the ground truth. CONCLUSIONS: To assess the performance of AI models in real-world clinical scenarios, we analyzed their performance in the ASFNR AI Competition. The first ASFNR Competition underscored the gap between expectation and reality; the models largely fell short in their assessments. As the integration of AI tools into clinical workflows increases, neuroradiologists must carefully recognize the capabilities, constraints, and consistency of these technologies. Before institutions adopt these algorithms, thorough validation is essential to ensure acceptable levels of performance in clinical settings.ABBREVIATIONS: AI = artificial intelligence; ASFNR = American Society of Functional Neuroradiology; AUROC = area under the receiver operating characteristic curve; DICOM = Digital Imaging and Communications in Medicine; GEE = generalized estimation equation; IQR = interquartile range; NPV = negative predictive value; PPV = positive predictive value; ROC = receiver operating characteristic; TBI = traumatic brain injury.

Model-based iterative reconstruction for direct imaging with point spread function encoded echo planar MRI.

BACKGROUND: Echo planar imaging (EPI) is a fast measurement technique commonly used in magnetic resonance imaging (MRI), but is highly sensitive to measurement non-idealities in reconstruction. Point spread function (PSF)-encoded EPI is a multi-shot strategy which alleviates distortion, but acquisition of encodings suitable for direct distortion-free imaging prolongs scan time. In this work, a model-based iterative reconstruction (MBIR) framework is introduced for direct imaging with PSF-EPI to improve image quality and acceleration potential. METHODS: An MBIR platform was developed for accelerated PSF-EPI. The reconstruction utilizes a subspace representation, is regularized to promote local low-rankedness (LLR), and uses variable splitting for efficient iteration. Comparisons were made against standard reconstructions from prospectively accelerated PSF-EPI data and with retrospective subsampling. Exploring aggressive partial Fourier acceleration of the PSF-encoding dimension, additional comparisons were made against an extension of Homodyne to direct PSF-EPI in numerical experiments. A neuroradiologists' assessment was completed comparing images reconstructed with MBIR from retrospectively truncated data directly against images obtained with standard reconstructions from non-truncated datasets. RESULTS: Image quality results were consistently superior for MBIR relative to standard and Homodyne reconstructions. As the MBIR signal model and reconstruction allow for arbitrary sampling of the PSF space, random sampling of the PSF-encoding dimension was also demonstrated, with quantitative assessments indicating best performance achieved through nonuniform PSF sampling combined with partial Fourier. With retrospective subsampling, MBIR reconstructs high-quality images from sub-minute scan datasets. MBIR was shown to be superior in a neuroradiologists' assessment with respect to three of five performance criteria, with equivalence for the remaining two. CONCLUSIONS: A novel image reconstruction framework is introduced for direct imaging with PSF-EPI, enabling arbitrary PSF space sampling and reconstruction of diagnostic-quality images from highly accelerated PSF-encoded EPI data.

7T MRI for Cushing Disease: A Single-Institution Experience and Literature Review.

BACKGROUND AND PURPOSE: Cushing disease is typically caused by a pituitary adenoma that frequently is small and challenging to detect on conventional MR imaging. High-field-strength 7T MR imaging can leverage increased SNR and contrast-to-noise ratios compared with lower-field-strength MR imaging to help identify small pituitary lesions. We aimed to describe our institutional experience with 7T MR imaging in patients with Cushing disease and perform a review of the literature. MATERIALS AND METHODS: We performed a retrospective analysis of 7T MR imaging findings in patients with pathology-proved Cushing disease from a single institution, followed by a review of the literature on 7T MR imaging for Cushing disease. RESULTS: Our institutional experience identified Cushing adenomas in 10/13 (76.9%) patients on 7T; however, only 5/13 (38.5%) lesions were discrete. Overall, the imaging protocols used were heterogeneous in terms of contrast dose as well as type of postcontrast T1-weighted sequences (dynamic, 2D versus 3D, and type of 3D sequence). From our institutional data, specific postgadolinium T1-weighted sequences were helpful in identifying a surgical lesion as follows: dynamic contrast-enhanced, 2/7 (28.6%); 2D FSE, 4/8 (50%); 3D sampling perfection with application-optimized contrasts by using different flip angle evolution (SPACE), 5/6 (83.3%); and 3D MPRAGE, 8/11 (72.7%). The literature review identified Cushing adenomas in 31/33 (93.9%) patients on 7T. CONCLUSIONS: 7T MR imaging for pituitary lesion localization in Cushing disease is a new technique with imaging protocols that vary widely. Further comparative research is needed to identify the optimal imaging technique as well as assess the benefit of 7T over lower-field-strength MR imaging.

Efficacy and safety of transcranial magnetic stimulation on cognition in mild cognitive impairment, Alzheimer's disease, Alzheimer's disease-related dementias, and other cognitive disorders: a systematic review and meta-analysis.

OBJECTIVE: We aim to analyze the efficacy and safety of TMS on cognition in mild cognitive impairment (MCI), Alzheimer's disease (AD), AD-related dementias, and nondementia conditions with comorbid cognitive impairment. DESIGN: Systematic review, Meta-Analysis. SETTING: We searched MEDLINE, Embase, Cochrane database, APA PsycINFO, Web of Science, and Scopus from January 1, 2000, to February 9, 2023. PARTICIPANTS AND INTERVENTIONS: RCTs, open-label, and case series studies reporting cognitive outcomes following TMS intervention were included. MEASUREMENT: Cognitive and safety outcomes were measured. Cochrane Risk of Bias for RCTs and MINORS (Methodological Index for Non-Randomized Studies) criteria were used to evaluate study quality. This study was registered with PROSPERO (CRD42022326423). RESULTS: The systematic review included 143 studies (n = 5,800 participants) worldwide, encompassing 94 RCTs, 43 open-label prospective, 3 open-label retrospective, and 3 case series. The meta-analysis included 25 RCTs in MCI and AD. Collectively, these studies provide evidence of improved global and specific cognitive measures with TMS across diagnostic groups. Only 2 studies (among 143) reported 4 adverse events of seizures: 3 were deemed TMS unrelated and another resolved with coil repositioning. Meta-analysis showed large effect sizes on global cognition (Mini-Mental State Examination (SMD = 0.80 [0.26, 1.33], p = 0.003), Montreal Cognitive Assessment (SMD = 0.85 [0.26, 1.44], p = 0.005), Alzheimer's Disease Assessment Scale-Cognitive Subscale (SMD = -0.96 [-1.32, -0.60], p < 0.001)) in MCI and AD, although with significant heterogeneity. CONCLUSION: The reviewed studies provide favorable evidence of improved cognition with TMS across all groups with cognitive impairment. TMS was safe and well tolerated with infrequent serious adverse events.

Detection and Mitigation of Neurovascular Uncoupling in Brain Gliomas.

Functional magnetic resonance imaging (fMRI) with blood oxygen level-dependent (BOLD) technique is useful for preoperative mapping of brain functional networks in tumor patients, providing reliable in vivo detection of eloquent cortex to help reduce the risk of postsurgical morbidity. BOLD task-based fMRI (tb-fMRI) is the most often used noninvasive method that can reliably map cortical networks, including those associated with sensorimotor, language, and visual functions. BOLD resting-state fMRI (rs-fMRI) is emerging as a promising ancillary tool for visualization of diverse functional networks. Although fMRI is a powerful tool that can be used as an adjunct for brain tumor surgery planning, it has some constraints that should be taken into consideration for proper clinical interpretation. BOLD fMRI interpretation may be limited by neurovascular uncoupling (NVU) induced by brain tumors. Cerebrovascular reactivity (CVR) mapping obtained using breath-hold methods is an effective method for evaluating NVU potential.

Improved Resting-State Functional MRI Using Multi-Echo Echo-Planar Imaging on a Compact 3T MRI Scanner with High-Performance Gradients.

In blood-oxygen-level-dependent (BOLD)-based resting-state functional (RS-fMRI) studies, usage of multi-echo echo-planar-imaging (ME-EPI) is limited due to unacceptable late echo times when high spatial resolution is used. Equipped with high-performance gradients, the compact 3T MRI system (C3T) enables a three-echo whole-brain ME-EPI protocol with smaller than 2.5 mm isotropic voxel and shorter than 1 s repetition time, as required in landmark fMRI studies. The performance of the ME-EPI was comprehensively evaluated with signal variance reduction and region-of-interest-, seed- and independent-component-analysis-based functional connectivity analyses and compared with a counterpart of single-echo EPI with the shortest TR possible. Through the multi-echo combination, the thermal noise level is reduced. Functional connectivity, as well as signal intensity, are recovered in the medial orbital sulcus and anterior transverse collateral sulcus in ME-EPI. It is demonstrated that ME-EPI provides superior sensitivity and accuracy for detecting functional connectivity and/or brain networks in comparison with single-echo EPI. In conclusion, the high-performance gradient enabled high-spatial-temporal resolution ME-EPI would be the method of choice for RS-fMRI study on the C3T.

Minimizing magnetic resonance image geometric distortion at 7 Tesla for frameless presurgical planning using skin-adhered fiducials.

BACKGROUND: 7T MRI offers significant benefits to spatial and contrast resolution compared to lower field strengths. This superior image quality can help better delineate targets in stereotactic neurosurgical procedures; however, the potential for increased geometric distortions at 7T has impaired its widespread use for these applications. Image geometric distortions can be due to distortions of B0 arising from tissue magnetic susceptibility effects or inherent field inhomogeneities, and nonlinearity of the magnetic field gradients. PURPOSE: The purpose of this study was to investigate the use of 7T MRI for neurosurgical frameless stereotactic navigation procedures. Image geometric distortions at the skin surface in 7T images were minimized and compared to results from clinical 3T frameless imaging protocols. METHODS: A 3D-printed grid phantom filled with oil was designed to perform a fine calibration of the 7T imaging gradients, and an oil-filled head phantom with internal targets was used to determine ground truth (from computed tomography [CT]) positioning errors. Three volunteers and the head phantom were imaged consecutively at 3T and 7T. Ten skin-adhesive fiducial markers were placed on each subject's exposed skin surface at standard clinical placement locations for frameless procedures. Imaging sequences included MPRAGE (three bandwidths at 7T: 400, 690, and 1020 Hz/pixel, and one at 3T: 400 Hz/pixel), T2 SPACE, and T2 SPACE FLAIR acquisitions. An additional GRE field map was acquired on both scanners using a multi-echo GRE sequence. Custom Matlab code was used to perform additional distortion correction of the images using the unwrapped field maps. Fiducial localization was performed with 3D Slicer, with absolute fiducial positioning errors determined in phantom experiments following rigid registration to the CT images. For human experiments, 3T and 7T images were registered and relative differences in fiducial locations were compared using two-tailed paired t-tests. RESULTS: Phantom measurements at 7T yielded gradient distance scaling errors of 1.1%, 2.2%, and 1.0% along the x-, y-, and z-axes, respectively. These system miscalibrations were traced back to phantom manufacturing deviations in the sphericity of the vendor's gradient calibration phantom. Correction factors along each gradient axis were applied, and afterward, geometric distortions of less than 1 mm were obtained in the 7T MR head phantom images for the 1020 Hz/pixel bandwidth MPRAGE sequence. For the human subjects, four fiducial locations were excluded from the analysis due to patient positioning differences. Differences between 3T and 7T MPRAGE with low/medium/high bandwidth were 2.2 /2.6/2.3 mm, respectively, before the correction, reducing to 1.6/1.3/1.0 mm after the correction (p < 0.001). T2 SPACE and T2 SPACE FLAIR yielded a similar pattern when the correction was applied, decreasing from 2.1 to 0.8 mm, and 2.6 to 1.0 mm, respectively. CONCLUSIONS: 7T MRI can be used to perform frameless presurgical planning with skin-adhesive fiducials. Geometric distortions can be reduced to a clinically relevant level (errors < âˆ¼1 mm) with no significant susceptibility-related distortions, by using high receiver bandwidth, ensuring gradients are properly calibrated, and placing skin fiducials in areas where distortions from patient positioning are minimal.

Enhanced clinical task-based fMRI metrics through locally low-rank denoising of complex-valued data.

OBJECTIVE: This study investigates a locally low-rank (LLR) denoising algorithm applied to source images from a clinical task-based functional MRI (fMRI) exam before post-processing for improving statistical confidence of task-based activation maps. METHODS: Task-based motor and language fMRI was obtained in eleven healthy volunteers under an IRB approved protocol. LLR denoising was then applied to raw complex-valued image data before fMRI processing. Activation maps generated from conventional non-denoised (control) data were compared with maps derived from LLR-denoised image data. Four board-certified neuroradiologists completed consensus assessment of activation maps; region-specific and aggregate motor and language consensus thresholds were then compared with nonparametric statistical tests. Additional evaluation included retrospective truncation of exam data without and with LLR denoising; a ROI-based analysis tracked t-statistics and temporal SNR (tSNR) as scan durations decreased. A test-retest assessment was performed; retest data were matched with initial test data and compared for one subject. RESULTS: fMRI activation maps generated from LLR-denoised data predominantly exhibited statistically significant (p = 4.88×10-4 to p = 0.042; one p = 0.062) increases in consensus t-statistic thresholds for motor and language activation maps. Following data truncation, LLR data showed task-specific increases in t-statistics and tSNR respectively exceeding 20 and 50% compared to control. LLR denoising enabled truncation of exam durations while preserving cluster volumes at fixed thresholds. Test-retest showed variable activation with LLR data thresholded higher in matching initial test data. CONCLUSION: LLR denoising affords robust increases in t-statistics on fMRI activation maps compared to routine processing, and offers potential for reduced scan duration while preserving map quality.

The development of ultra-high field MRI guidance technology for neuronavigation.

OBJECTIVE: Magnetic resonance imaging at 7T offers improved image spatial and contrast resolution for visualization of small brain nuclei targeted in neuromodulation. However, greater image geometric distortion and a lack of compatible instrumentation preclude implementation. In this report, the authors detail the development of a stereotactic image localizer and accompanying imaging sequences designed to mitigate geometric distortion, enabling accurate image registration and surgical planning of basal ganglia nuclei. METHODS: Magnetization-prepared rapid acquisition with gradient echo (MPRAGE), fast gray matter acquisition T1 inversion recovery (FGATIR), T2-weighted, and T2*-weighted sequences were optimized for 7T in 9 human subjects to visualize basal ganglia nuclei, minimize image distortion, and maximize target contrast-to-noise and signal-to-noise ratios. Extracranial spatial distortions were mapped to develop a skull-contoured image localizer embedded with spherical silicone fiducials for improved MR image registration and target guidance. Surgical plan accuracy testing was initially performed in a custom-developed MRI phantom (n = 5 phantom studies) and finally in a human trial. RESULTS: MPRAGE and T2*-weighted sequences had the best measures among global measures of image quality (3.8/4, p < 0.0001; and 3.7/4, p = 0.0002, respectively). Among basal ganglia nuclei, FGATIR outperformed MPRAGE for globus pallidus externus (GPe) visualization (2.67/4 vs 1.78/4, p = 0.008), and FGATIR, T2-weighted imaging, and T2*-weighted imaging outperformed MPRAGE for substantia nigra visualization (1.44/4 vs 2.56/4, p = 0.04; vs 2.56/4, p = 0.04; vs 2.67/4, p = 0.003). Extracranial distortion was lower in the head's midregion compared with the base and apex ( 1.17-1.33 mm; MPRAGE and FGATIR, p < 0.0001; T2-weighted imaging, p > 0.05; and T2*-weighted imaging, p = 0.013). Fiducial placement on the localizer in low distortion areas improved image registration (fiducial registration error, 0.79-1.19 mm; p < 0.0001) and targeting accuracy (target registration error, 0.60-1.09 mm; p = 0.04). Custom surgical software and the refined image localizer enabled successful surgical planning in a human trial (fiducial registration error = 1.0 mm). CONCLUSIONS: A skull-contoured image localizer that accounts for image distortion is necessary to enable high-accuracy 7T imaging-guided targeting for surgical neuromodulation. These results may enable improved clinical efficacy for the treatment of neurological disease.

Implementation and Rationale for a Unified Clinical and Imaging Protocol for Evaluation and Treatment of Moyamoya Angiopathy: A Single Institutional Experience.

Moyamoya disease (MMD) is a complex and incompletely-understood cerebrovascular pathological entity that requires thorough clinical and imaging evaluation. Moyamoya is rare, thereby making the establishment of an effective, thorough and interdisciplinary patient evaluation protocol challenging, even within specialized referral centers. Nevertheless, implementation of such a protocol is crucial in order to provide the best possible evaluation and treatment for MMD patients. Here, we describe our institution's implementation of, rationale for, and experience with a comprehensive multidisciplinary collaboration and evaluation strategy for adult patients with moyamoya. This evaluation course consists of, first of all, a thorough clinical and laboratory evaluation with a vascular neurologist. This is followed by a comprehensive imaging assessment which evaluates angiographic and parenchymal features, in addition to cerebrovascular functionality. Finally, appropriate referrals are made to consulting services as indicated, which includes vascular neurosurgery. These steps are described in detail herein.

Clinical 7-T MRI for neuroradiology: strengths, weaknesses, and ongoing challenges.

Since the relatively recent regulatory approval for clinical use in both Europe and North America, 7-Tesla (T) MRI has been adopted for clinical practice at our institution. Based on this experience, this article reviews the unique features of 7-T MRI neuroimaging and addresses the challenges of establishing a 7-T MRI clinical practice. The underlying fundamental physics principals of high-field strength MRI are briefly reviewed. Scanner installation, safety considerations, and artifact mitigation techniques are discussed. Seven-tesla MRI case examples of neurologic diseases including epilepsy, vascular abnormalities, and tumor imaging are presented to illustrate specific applications of 7-T MRI. The advantages of 7-T MRI in conjunction with advanced neuroimaging techniques such as functional MRI are presented. Seven-tesla MRI produces more detailed information and, in some cases, results in specific diagnoses where previous 3-T studies were insufficient. Still, persistent technical issues for 7-T scanning present ongoing challenges for radiologists.

Higher temporal resolution multiband fMRI provides improved presurgical language maps.

PURPOSE: We investigated the hypothesis that increasing fMRI temporal resolution using a multiband (MB) gradient echo-echo planar imaging (GRE-EPI) pulse sequence provides fMRI language maps of higher statistical quality than those acquired with a traditional GRE-EPI sequence. METHODS: This prospective study enrolled 29 consecutive patients receiving language fMRI prior to a potential brain resection for tumor, AVM, or epilepsy. A 4-min rhyming task was performed at 3.0 Tesla with a traditional GRE-EPI pulse sequence (TR = 2000, TE = 30, matrix = 64/100%, slice = 4/0, FOV = 24, slices = 30, time points = 120) and an additional MB GRE-EPI pulse sequence with an acceleration factor of 6 (TR = 333, TE = 30, matrix 64/100%, slice = 4/0, FOV = 24, time points = 720). Spatially filtered t statistical maps were generated. Volumes of interest (VOIs) were drawn around activations at Broca's, dorsolateral prefrontal cortex, Wernicke's, and the visual word form areas. The t value maxima were measured for the overall brain and each of the VOIs. A paired t test was performed for the corresponding traditional and MB GRE-EPI measurements. RESULTS: The mean age of subjects was 42.6 years old (18-75). Sixty-two percent were male. The average overall brain t statistic maxima for the MB pulse sequence (t = 15.4) was higher than for the traditional pulse sequence (t = 9.3, p = < .0001). This also held true for Broca's area (p < 0.0001), Wernicke's area (p < .0001), dorsolateral prefrontal cortex (p < .0001), and the visual word form area (p < .0001). CONCLUSION: A MB GRE-EPI fMRI pulse sequence employing high temporal resolution provides clinical fMRI language maps of greater statistical significance than those obtained with a traditional GRE-EPI sequence.

Magnetic Resonance Safety in the 7T Environment.

The arrival of 7T MR imaging into the clinic represents a significant step-change in MR technology. This article describes safety concerns associated with imaging at 7T, including the increased magnetic forces on magnetic objects at 7T and the interaction of the 300 MHz (Larmor) radiofrequency energy with tissue in the body. A dedicated multidisciplinary 7T Safety team should develop safety policies and procedures to address these safety challenges and keep abreast of best practice in the field. The off-label imaging of implanted devices is discussed, and also the need for staff training to deal with complexities of patient handling and image interpretation.

Current concepts of cross-sectional and functional anatomy of the cerebellum: a pictorial review and atlas.

Recognition of key concepts of structural and functional anatomy of the cerebellum can facilitate image interpretation and clinical correlation. Recently, the human brain mapping literature has increased our understanding of cerebellar anatomy, function, connectivity with the cerebrum, and significance of lesions involving specific areas.Both the common names and numerically based Schmahmann classifications of cerebellar lobules are illustrated. Anatomic patterns, or signs, of key fissures and white matter branching are introduced to facilitate easy recognition of the major anatomic features. Color-coded overlays of cross-sectional imaging are provided for reference of more complex detail. Examples of exquisite detail of structural and functional cerebellar anatomy at 7 T MRI are also depicted.The functions of the cerebellum are manifold with the majority of areas involved with non-motor association function. Key concepts of lesion-symptom mapping which correlates lesion location to clinical manifestation are introduced, emphasizing that lesions in most areas of the cerebellum are associated with predominantly non-motor deficits. Clinical correlation is reinforced with examples of intrinsic pathologic derangement of cerebellar anatomy and altered functional connectivity due to pathology of the cerebral hemisphere. The purpose of this pictorial review is to illustrate basic concepts of these topics in a cross-sectional imaging-based format that can be easily understood and applied by radiologists.

Image Artifact Management for Clinical Magnetic Resonance Imaging on a 7 T Scanner Using Single-Channel Radiofrequency Transmit Mode.

OBJECTIVES: The aim of this work was to devise mitigation strategies for addressing a range of image artifacts on a clinical 7 T magnetic resonance imaging scanner using the regulatory-approved single-channel radiofrequency transmit mode and vendor-supplied radiofrequency coils to facilitate clinical scanning within reasonable scan times. MATERIALS AND METHODS: Optimized imaging sequence protocols were developed for routine musculoskeletal knee and neurological imaging. Sources of severe image nonuniformities were identified, and mitigation strategies were devised. A range of custom-made high permittivity dielectric pads were used to compensate for B1 and B1 inhomogeneities, and also for magnetic susceptibility-induced signal dropouts particularly in the basal regions of the temporal lobes and in the cerebellum. RESULTS: Significant improvements in image uniformity were obtained using dielectric pads in the knee and brain. A combination of small voxels, reduced field of view B0 shimming, and high in-plane parallel imaging factors helped to minimize signal loss in areas of high susceptibility-induced field distortions. The high inherent signal-to-noise ratio at 7 T allowed for high receiver bandwidths and thin slices to minimize chemical shift artifacts. Intermittent artifacts due to radiofrequency inversion pulse limitations (power, bandwidth) were minimized with dielectric pads. A patient with 2 implanted metallic cranial fixation devices located within the radiofrequency transmit field was successfully imaged, with minimal image geometric distortions. CONCLUSIONS: Challenges relating to severe image artifacts at 7 T using single-channel radiofrequency transmit functionality in the knee and brain were overcome using the approaches described in this article. The resultant high diagnostic image quality paves the way for incorporation of this technology into the routine clinical workflow. Further developmental efforts are required to expand the range of applications to other anatomical areas, and to expand the evidence- and knowledge-base relating to the safety of scanning patients with implanted metallic devices.

Segmentation errors and intertest reliability in automated and manually traced hippocampal volumes.

OBJECTIVE: To rigorously compare automated atlas-based and manual tracing hippocampal segmentation for accuracy, repeatability, and clinical acceptability given a relevant range of imaging abnormalities in clinical epilepsy. METHODS: Forty-nine patients with hippocampal asymmetry were identified from our institutional radiology database, including two patients with significant anatomic deformations. Manual hippocampal tracing was performed by experienced technologists on 3T MPRAGE images, measuring hippocampal volume up to the tectal plate, excluding the hippocampal tail. The same images were processed using NeuroQuant and FreeSurfer software. Ten subjects underwent repeated manual hippocampal tracings by two additional technologists blinded to previous results to evaluate consistency. Ten patients with two clinical MRI studies had volume measurements repeated using NeuroQuant and FreeSurfer. RESULTS: FreeSurfer raw volumes were significantly lower than NeuroQuant (P < 0.001, right and left), and hippocampal asymmetry estimates were lower for both automatic methods than manual tracing (P < 0.0001). Differences remained significant after scaling volumes to age, gender, and scanner matched normative percentiles. Volume reproducibility was fair (0.4-0.59) for manual tracing, and excellent (>0.75) for both automated methods. Asymmetry index reproducibility was excellent (>0.75) for manual tracing and FreeSurfer segmentation and fair (0.4-0.59) for NeuroQuant segmentation. Both automatic segmentation methods failed on the two cases with anatomic deformations. Segmentation errors were visually identified in 25 NeuroQuant and 27 FreeSurfer segmentations, and nine (18%) NeuroQuant and six (12%) FreeSurfer errors were judged clinically significant. INTERPRETATION: Automated hippocampal volumes are more reproducible than hand-traced hippocampal volumes. However, these methods fail in some cases, and significant segmentation errors can occur.

Relationship Between Seizure Frequency and Functional Abnormalities in Limbic Network of Medial Temporal Lobe Epilepsy.

Background: We compared resting-state functional connectivity (RSFC) among limbic and temporal lobe regions between patients with medial temporal lobe epilepsy (mTLE) and healthy control subjects to identify imaging evidence of functional networks related to seizure frequency, age of seizure onset, and duration of epilepsy. Methods: Twelve patients with drug-resistant, unilateral medial temporal lobe epilepsy and 12 healthy control subjects matched for age, sex, and handedness participated in the imaging experiments. We used network-based statistics to compare functional connectivity graphs in patients with mTLE and healthy controls to investigate the relationship between functional connectivity abnormalities and seizure frequency. Results: Among mTLE patients, we found functional network abnormalities throughout the limbic system, but primarily in the hemisphere ipsilateral to the seizure focus. The RSFCs between ipsilateral hypothalamus and ventral anterior cingulate cortex and between ipsilateral subiculum and contralateral posterior cingulate cortex were highly correlated with seizure frequency. Discussion: These findings suggest that in mTLE, changes in limbic networks ipsilateral to the epileptic focus are common. The pathological changes in connectivity between cingulate cortex, hypothalamus and subiculum ipsilateral to the seizure focus were correlated with increased seizure frequency.

Nuclei-specific thalamic connectivity predicts seizure frequency in drug-resistant medial temporal lobe epilepsy.

BACKGROUND AND OBJECTIVES: We assessed correlations between the resting state functional connectivity (RSFC) of different thalamic nuclei and seizure frequency in patients with drug-resistant medial temporal lobe epilepsy (mTLE). METHODS: Seventeen patients with mTLE and 17 sex-/age-/handedness-matched controls participated. A seed-based correlation method for the resting-state FMRI data was implemented to get RSFC maps of 70 thalamic nuclei seed masks. Group statistics for individual RSFC for subjects and seed masks were performed to obtain within-group characteristics and between-group differences with age covariates. A linear regression was applied to test whether seizure frequency correlated with thalamic nuclear RSFC with the whole brain in mTLE patients. RESULTS: RSFC of thalamic nuclei showed spatially distinguishable connectivity patterns that reflected principal inputs and outputs that were derived from priori anatomical knowledge. We found group differences between normal control and mTLE groups in RSFC for nuclei seeds located in various subdivisions of thalamus. The RSFCs in some of those nuclei were strongly correlated with seizure frequency. CONCLUSIONS: Mediodorsal thalamic nuclei may play important roles in seizure activity or in the regulation of neuronal activity in the limbic system. The RSFC of motor- and sensory-relay nuclei may help elucidate sensory-motor deficits associated with chronic seizure activity. RSFC of the pulvinar nuclei of the thalamus could also be a key reflection of symptom-related functional deficits in mTLE.