Automated segmentation of basal ganglia and deep brain structures in MRI of Parkinson's disease

Improved segmentation of basal ganglia from MR images using convolutional neural network with crossover-typed skip connection

PURPOSE

Accurate and automatic segmentation of basal ganglia from magnetic resonance (MR) images is important for diagnosis and treatment of various brain disorders. However, the basal ganglia segmentation is a challenging task because of the class imbalance and the unclear boundaries among basal ganglia anatomical structures. Thus, we aim to present an encoder-decoder convolutional neural network (CNN)-based method for improved segmentation of basal ganglia by focusing on skip connections that determine the segmentation performance of encoder-decoder CNNs. We also aim to reveal the effect of skip connections on the segmentation of basal ganglia with unclear boundaries.

METHODS

We used the encoder-decoder CNNs with the following five patterns of skip connections: without skip connection, with full-resolution horizontal skip connection, with horizontal skip connections, with vertical skip connections, and with crossover-typed skip connections (the proposed method). We compared and evaluated the performance of the CNNs in the experiment of basal ganglia segmentation using T1-weighted MR brain images of 79 patients.

RESULTS

The experimental results showed that the skip connections at each scale level help CNNs to acquire multi-scale image features, the vertical skip connections contribute on acquiring finer image features for segmentation of smaller anatomical structures with more blurred boundaries, and the crossover-typed skip connections, a combination of horizontal and vertical skip connections, provided better segmentation accuracy.

CONCLUSION

This paper investigated the effect of skip connections on the basal ganglia segmentation and revealed the crossover-typed skip connections might be effective for improving the segmentation of basal ganglia with the class imbalance and the unclear boundaries.

A preliminary exploration into top-down and bottom-up deep-learning approaches to localising neuro-interventional point targets in volumetric MRI

PURPOSE

Point localisation is a critical aspect of many interventional planning procedures, specifically representing anatomical regions of interest or landmarks as individual points. This could be seen as analogous to the problem of visual search in cognitive psychology, in which this search is performed either: bottom-up, constructing increasingly abstract and coarse-resolution features over the entire image; or top-down, using contextual cues from the entire image to refine the scope of the region being investigated. Traditional convolutional neural networks use the former, but it is not clear if this is optimal. This article is a preliminary investigation as to how this motivation affects 3D point localisation in neuro-interventional planning.

METHODS

Two neuro-imaging datasets were collected: one for cortical point localisation for repetitive transcranial magnetic stimulation and the other for sub-cortical anatomy localisation for deep brain stimulation. Four different frameworks were developed using top-down versus bottom-up paradigms as well as representing points as co-ordinates or heatmaps. These networks were applied to point localisation for transcranial magnetic stimulation and subcortical anatomy localisation. These networks were evaluated using cross-validation and a varying number of training datasets to analyse their sensitivity to quantity of training data.

RESULTS

Each network shows increasing performance as the amount of available training data increases, with the co-ordinate-based top-down network consistently outperforming the others. Specifically, the top-down architectures tend to outperform the bottom-up ones. An analysis of their memory consumption also encourages the top-down co-ordinate based architecture as it requires significantly less memory than either bottom-up architectures or those representing their predictions via heatmaps.

CONCLUSION

This paper is a preliminary foray into a fundamental aspect of machine learning architectural design: that of the top-down/bottom-up divide from cognitive psychology. Although there are additional considerations within the particular architectures investigated that could affect these results and the number of architectures investigated is limited, our results do indicate that the less commonly used top-down paradigm could lead to more efficient and effective architectures in the future.

Neuroimaging and verbal memory assessment in healthy aging adults using a portable low-field MRI scanner and a web-based platform: results from a proof-of-concept population-based cross-section study

Consumer wearables and health monitors, internet-based health and cognitive assessments, and at-home biosample (e.g., saliva and capillary blood) collection kits are increasingly used by public health researchers for large population-based studies without requiring intensive in-person visits. Alongside reduced participant time burden, remote and virtual data collection allows the participation of individuals who live long distances from hospital or university research centers, or who lack access to transportation. Unfortunately, studies that include magnetic resonance neuroimaging are challenging to perform remotely given the infrastructure requirements of MRI scanners, and, as a result, they often omit socially, economically, and educationally disadvantaged individuals. Lower field strength systems (< 100 mT) offer the potential to perform neuroimaging at a participant's home, enabling more accessible and equitable research. Here we report the first use of a low-field MRI "scan van" with an online assessment of paired-associate learning (PAL) to examine associations between brain morphometry and verbal memory performance. In a sample of 67 individuals, 18-93 years of age, imaged at or near their home, we show expected white and gray matter volume trends with age and find significant (p < 0.05 FWE) associations between PAL performance and hippocampus, amygdala, caudate, and thalamic volumes. High-quality data were acquired in 93% of individuals, and at-home scanning was preferred by all individuals with prior MRI at a hospital or research setting. Results demonstrate the feasibility of remote neuroimaging and cognitive data collection, with important implications for engaging traditionally under-represented communities in neuroimaging research.

High-resolution atlasing and segmentation of the subcortex: Review and perspective on challenges and opportunities created by machine learning

This paper reviews almost three decades of work on atlasing and segmentation methods for subcortical structures in human brain MRI. In writing this survey, we have three distinct aims. First, to document the evolution of digital subcortical atlases of the human brain, from the early MRI templates published in the nineties, to the complex multi-modal atlases at the subregion level that are available today. Second, to provide a detailed record of related efforts in the automated segmentation front, from earlier atlas-based methods to modern machine learning approaches. And third, to present a perspective on the future of high-resolution atlasing and segmentation of subcortical structures in in vivo human brain MRI, including open challenges and opportunities created by recent developments in machine learning.

Voxel-based diktiometry: Combining convolutional neural networks with voxel-based analysis and its application in diffusion tensor imaging for Parkinson's disease

Extracting population‐wise information from medical images, specifically in the neurological domain, is crucial to better understanding disease processes and progression. This is frequently done in a whole‐brain voxel‐wise manner, in which a population of patients and healthy controls are registered to a common co‐ordinate space and a statistical test is performed on the distribution of image intensities for each location. Although this method has yielded a number of scientific insights, it is further from clinical applicability as the differences are often small and altogether do not permit for a high‐performing classifier. In this article, we take the opposite approach of using a high‐performing classifier, specifically a traditional convolutional neural network, and then extracting insights from it which can be applied in a population‐wise manner, a method we call voxel‐based diktiometry. We have applied this method to diffusion tensor imaging (DTI) analysis for Parkinson's disease (PD), using the Parkinson's Progression Markers Initiative database. By using the network sensitivity information, we can decompose what elements of the DTI contribute the most to the network's performance, drawing conclusions about diffusion biomarkers for PD that are based on metrics which are not readily expressed in the voxel‐wise approach.

Combining simple interactivity and machine learning: a separable deep learning approach to subthalamic nucleus localization and segmentation in MRI for deep brain stimulation surgical planning

Purpose: Deep brain stimulation (DBS) is an interventional treatment for some neurological and neurodegenerative diseases. For example, in Parkinson's disease, DBS electrodes are positioned at particular locations within the basal ganglia to alleviate the patient's motor symptoms. These interventions depend greatly on a preoperative planning stage in which potential targets and electrode trajectories are identified in a preoperative MRI. Due to the small size and low contrast of targets such as the subthalamic nucleus (STN), their segmentation is a difficult task. Machine learning provides a potential avenue for development, but it has difficulty in segmenting such small structures in volumetric images due to additional problems such as segmentation class imbalance. Approach: We present a two-stage separable learning workflow for STN segmentation consisting of a localization step that detects the STN and crops the image to a small region and a segmentation step that delineates the structure within that region. The goal of this decoupling is to improve accuracy and efficiency and to provide an intermediate representation that can be easily corrected by a clinical user. This correction capability was then studied through a human-computer interaction experiment with seven novice participants and one expert neurosurgeon. Results: Our two-step segmentation significantly outperforms the comparative registration-based method currently used in clinic and approaches the fundamental limit on variability due to the image resolution. In addition, the human-computer interaction experiment shows that the additional interaction mechanism allowed by separating STN segmentation into two steps significantly improves the users' ability to correct errors and further improves performance. Conclusions: Our method shows that separable learning not only is feasible for fully automatic STN segmentation but also leads to improved interactivity that can ease its translation into clinical use.

An ecological approach to identify distinct neural correlates of disinhibition in frontotemporal dementia

Disinhibition is a core symptom of many neurodegenerative diseases, particularly frontotemporal dementia, and is a major cause of stress for caregivers. While a distinction between behavioural and cognitive disinhibition is common, an operational definition of behavioural disinhibition is still missing. Furthermore, conventional assessment of behavioural disinhibition, based on questionnaires completed by the caregivers, often lacks ecological validity. Therefore, their neuroanatomical correlates are non-univocal. In the present work, we used an original behavioural approach in a semi-ecological situation to assess two specific dimensions of behavioural disinhibition: compulsivity and social disinhibition. First, we investigated disinhibition profile in patients compared to controls. Then, to validate our approach, compulsivity and social disinhibition scores were correlated with classic cognitive tests measuring disinhibition (Hayling Test) and social cognition (mini-Social cognition & Emotional Assessment). Finally, we disentangled the anatomical networks underlying these two subtypes of behavioural disinhibition, taking in account the grey (voxel-based morphometry) and white matter (diffusion tensor imaging tractography). We included 17 behavioural variant frontotemporal dementia patients and 18 healthy controls. We identified patients as more compulsive and socially disinhibited than controls. We found that behavioural metrics in the semi-ecological task were related to cognitive performance: compulsivity correlated with the Hayling test and both compulsivity and social disinhibition were associated with the emotion recognition test. Based on voxel-based morphometry and tractography, compulsivity correlated with atrophy in the bilateral orbitofrontal cortex, the right temporal region and subcortical structures, as well as with alterations of the bilateral cingulum and uncinate fasciculus, the right inferior longitudinal fasciculus and the right arcuate fasciculus. Thus, the network of regions related to compulsivity matched the "semantic appraisal" network. Social disinhibition was associated with bilateral frontal atrophy and impairments in the forceps minor, the bilateral cingulum and the left uncinate fasciculus, regions corresponding to the frontal component of the "salience" network. Summarizing, this study validates our semi-ecological approach, through the identification of two subtypes of behavioural disinhibition, and highlights different neural networks underlying compulsivity and social disinhibition. Taken together, these findings are promising for clinical practice by providing a better characterisation of inhibition disorders, promoting their detection and consequently a more adapted management of patients.

Past, Present, and Future of Deep Brain Stimulation: Hardware, Software, Imaging, Physiology and Novel Approaches

Deep brain stimulation (DBS) has advanced treatment options for a variety of neurologic and neuropsychiatric conditions. As the technology for DBS continues to progress, treatment efficacy will continue to improve and disease indications will expand. Hardware advances such as longer-lasting batteries will reduce the frequency of battery replacement and segmented leads will facilitate improvements in the effectiveness of stimulation and have the potential to minimize stimulation side effects. Targeting advances such as specialized imaging sequences and "connectomics" will facilitate improved accuracy for lead positioning and trajectory planning. Software advances such as closed-loop stimulation and remote programming will enable DBS to be a more personalized and accessible technology. The future of DBS continues to be promising and holds the potential to further improve quality of life. In this review we will address the past, present and future of DBS.

Automatic segmentation of Deep Grey Nuclei using a high-resolution 7T MRI Atlas - quantification of T1 values in healthy volunteers

We present a new consensus atlas of deep grey nuclei obtained by shape-based averaging of manual segmentation of two experienced neuroradiologists and optimized from 7T MP2RAGE images acquired at (0.6mm)³ in 60 healthy subjects. A group-wise normalization method was used to build a high-contrast and high-resolution T₁ -weighted brain template (0.5mm)³ using data from 30 out of the 60 controls. Delineation of 24 deep grey nuclei per hemisphere, including the claustrum and twelve thalamic nuclei, was then performed by two expert neuroradiologists and reviewed by a third neuroradiologist according to tissue contrast and external references based on the Morel atlas. Corresponding deep grey matter structures were also extracted from the Morel and CIT168 atlases. The data-derived, Morel and CIT168 atlases were all applied at the individual level using non-linear registration to fit the subject reference and to extract absolute mean quantitative T₁ values derived from the 3D-MP2RAGE volumes, after correction for residual B₁ ⁺ biases. Three metrics (The Dice and the volumetric similarity coefficients, and a novel Hausdorff distance) were used to estimate the inter-rater agreement of manual MRI segmentation and inter-atlas variability, and these metrics were measured to quantify biases due to image registration and their impact on the measurements of the quantitative T₁ values was highlighted. This represents a fully-automated segmentation process permitting the extraction of unbiased normative T₁ values in a population of young healthy controls as a reference for characterizing subtle structural alterations of deep grey nuclei relevant to a range of neurological diseases.

Automated segmentation of deep brain nuclei using convolutional neural networks and susceptibility weighted imaging

The advent of susceptibility-sensitive MRI techniques, such as susceptibility weighted imaging (SWI), has enabled accurate in vivo visualization and quantification of iron deposition within the human brain. Although previous approaches have been introduced to segment iron-rich brain regions, such as the substantia nigra, subthalamic nucleus, red nucleus, and dentate nucleus, these methods are largely unavailable and manual annotation remains the most used approach to label these regions. Furthermore, given their recent success in outperforming other segmentation approaches, convolutional neural networks (CNN) promise better performances. The aim of this study was thus to evaluate state-of-the-art CNN architectures for the labeling of deep brain nuclei from SW images. We implemented five CNN architectures and considered ensembles of these models. Furthermore, a multi-atlas segmentation model was included to provide a comparison not based on CNN. We evaluated two prediction strategies: individual prediction, where a model is trained independently for each region, and combined prediction, which simultaneously predicts multiple closely located regions. In the training dataset, all models performed with high accuracy with Dice coefficients ranging from 0.80 to 0.95. The regional SWI intensities and volumes from the models' labels were strongly correlated with those obtained from manual labels. Performances were reduced on the external dataset, but were higher or comparable to the intrarater reliability and most models achieved significantly better results compared to multi-atlas segmentation. CNNs can accurately capture the individual variability of deep brain nuclei and represent a highly useful tool for their segmentation from SW images.

Clustering of Brain Tumor Based on Analysis of MRI Images Using Robust Principal Component Analysis (ROBPCA) Algorithm

Automated detection of brain tumor location is essential for both medical and analytical uses. In this paper, we clustered brain MRI images to detect tumor location. To obtain perfect results, we presented an unsupervised robust PCA algorithm to clustered images. The proposed method clusters brain MR image pixels to four leverages. The algorithm is implemented for five brain diseases such as glioma, Huntington, meningioma, Pick, and Alzheimer's. We used ten images of each disease to validate the optimal identification rate. According to the results obtained, 2% of the data in the bad leverage part of the image were determined, which acceptably discerned the tumor. Results show that this method has the potential to detect tumor location for brain disease with high sensitivity. Moreover, results show that the method for the Glioma images has approximately better results than others. However, according to the ROC curve for all selected diseases, the present method can find lesion location.

Image Guidance in Deep Brain Stimulation Surgery to Treat Parkinson's Disease: A Comprehensive Review

Deep brain stimulation (DBS) is an effective therapy as an alternative to pharmaceutical treatments for Parkinson's disease (PD). Aside from factors such as instrumentation, treatment plans, and surgical protocols, the success of the procedure depends heavily on the accurate placement of the electrode within the optimal therapeutic targets while avoiding vital structures that can cause surgical complications and adverse neurologic effects. Although specific surgical techniques for DBS can vary, interventional guidance with medical imaging has greatly contributed to the development, outcomes, and safety of the procedure. With rapid development in novel imaging techniques, computational methods, and surgical navigation software, as well as growing insights into the disease and mechanism of action of DBS, modern image guidance is expected to further enhance the capacity and efficacy of the procedure in treating PD. This article surveys the state-of-the-art techniques in image-guided DBS surgery to treat PD, and discusses their benefits and drawbacks, as well as future directions on the topic.

pBrain: A novel pipeline for Parkinson related brain structure segmentation

•

A novel pipeline for Parkinson`s disease structure segmentation is presented.

•

State-of-the-art fast multiatlas patch-based label fusion with systematic error correction is used to accurately and efficiently produce very competitive results in around 5 min.

•

The proposed pipeline works at high resolution (0.5 mm) but it can work also with standard resolution (1 mm) T2 images allowing the analysis of large legacy databases.

•

The proposed pipeline will be made publically available online through our volBrain platform.

Parkinson is a very prevalent neurodegenerative disease impacting the life of millions of people worldwide. Although its cause remains unknown, its functional and structural analysis is fundamental to advance in the search of a cure or symptomatic treatment. The automatic segmentation of deep brain structures related to Parkinson`s disease could be beneficial for the follow up and treatment planning. Unfortunately, there is not broadly available segmentation software to automatically measure Parkinson related structures. In this paper, we present a novel pipeline to segment three deep brain structures related to Parkinson's disease (substantia nigra, subthalamic nucleus and red nucleus). The proposed method is based on the multi-atlas label fusion technology that works on standard and high-resolution T2-weighted images. The proposed method also includes as post-processing a new neural network-based error correction step to minimize systematic segmentation errors. The proposed method has been compared to other state-of-the-art methods showing competitive results in terms of accuracy and execution time.

Impact of CT reconstruction algorithm on auto-segmentation performance

Model‐based iterative reconstruction (MBIR) reduces CT imaging dose while maintaining image quality. However, MBIR reduces noise while preserving edges which may impact intensity‐based tasks such as auto‐segmentation. This work evaluates the sensitivity of an auto‐contouring prostate atlas across multiple MBIR reconstruction protocols and benchmarks the results against filtered back projection (FBP). Images were created from raw projection data for 11 prostate cancer cases using FBP and nine different MBIR reconstructions (3 protocols/3 noise reduction levels) yielding 10 reconstructions/patient. Five bony structures, bladder, rectum, prostate, and seminal vesicles (SVs) were segmented using an auto‐segmentation pipeline that renders 3D binary masks for analysis. Performance was evaluated for volume percent difference (VPD) and Dice similarity coefficient (DSC), using FBP as the gold standard. Nonparametric Friedman tests plus post hoc all pairwise comparisons were employed to test for significant differences (P < 0.05) for soft tissue organs and protocol/level combinations. A physician performed qualitative grading of 396 MBIR contours across the prostate, bladder, SVs, and rectum in comparison to FBP using a six‐point scale. MBIR contours agreed with FBP for bony anatomy (DSC ≥ 0.98), bladder (DSC ≥ 0.94, VPD < 8.5%), and prostate (DSC = 0.94 ± 0.03, VPD = 4.50 ± 4.77% (range: 0.07–26.39%). Increased variability was observed for rectum (VPD = 7.50 ± 7.56% and DSC = 0.90 ± 0.08) and SVs (VPD and DSC of 8.23 ± 9.86% range (0.00–35.80%) and 0.87 ± 0.11, respectively). Over the all protocol/level comparisons, a significant difference was observed for the prostate VPD between BSPL1 and BSTL2 (adjusted P‐value = 0.039). Nevertheless, 300 of 396 (75.8%) of the four soft tissue structures using MBIR were graded as equivalent or better than FBP, suggesting that MBIR offered potential improvements in auto‐segmentation performance when compared to FBP. Future work may involve tuning organ‐specific MBIR parameters to further improve auto‐segmentation performance. Running title: Impact of CT Reconstruction Algorithm on Auto‐segmentation Performance.

Multi-modal imaging with specialized sequences improves accuracy of the automated subcortical grey matter segmentation

The basal ganglia and limbic system, particularly the thalamus, putamen, internal and external globus pallidus, substantia nigra, and sub-thalamic nucleus, comprise a clinically relevant signal network for Parkinson's disease. In order to manually trace these structures, a combination of high-resolution and specialized sequences at 7 T are used, but it is not feasible to routinely scan clinical patients in those scanners. Targeted imaging sequences at 3 T have been presented to enhance contrast in a select group of these structures. In this work, we show that a series of atlases generated at 7 T can be used to accurately segment these structures at 3 T using a combination of standard and optimized imaging sequences, though no one approach provided the best result across all structures. In the thalamus and putamen, a median Dice Similarity Coefficient (DSC) over 0.88 and a mean surface distance <1.0 mm were achieved using a combination of T1 and an optimized inversion recovery imaging sequences. In the internal and external globus pallidus a DSC over 0.75 and a mean surface distance <1.2 mm were achieved using a combination of T1 and inversion recovery imaging sequences. In the substantia nigra and sub-thalamic nucleus a DSC of over 0.6 and a mean surface distance of <1.0 mm were achieved using the inversion recovery imaging sequence. On average, using T1 and optimized inversion recovery together significantly improved segmentation results than over individual modality (p < 0.05 Wilcoxon sign-rank test).

Automatic localization of the subthalamic nucleus on patient-specific clinical MRI by incorporating 7 T MRI and machine learning: Application in deep brain stimulation

Deep brain stimulation (DBS) of the subthalamic nucleus (STN) has shown clinical potential for relieving the motor symptoms of advanced Parkinson's disease. While accurate localization of the STN is critical for consistent across-patients effective DBS, clear visualization of the STN under standard clinical MR protocols is still challenging. Therefore, intraoperative microelectrode recordings (MER) are incorporated to accurately localize the STN. However, MER require significant neurosurgical expertise and lengthen the surgery time. Recent advances in 7 T MR technology facilitate the ability to clearly visualize the STN. The vast majority of centers, however, still do not have 7 T MRI systems, and fewer have the ability to collect and analyze the data. This work introduces an automatic STN localization framework based on standard clinical MRIs without additional cost in the current DBS planning protocol. Our approach benefits from a large database of 7 T MRI and its clinical MRI pairs. We first model in the 7 T database, using efficient machine learning algorithms, the spatial and geometric dependency between the STN and its adjacent structures (predictors). Given a standard clinical MRI, our method automatically computes the predictors and uses the learned information to predict the patient-specific STN. We validate our proposed method on clinical T2 W MRI of 80 subjects, comparing with experts-segmented STNs from the corresponding 7 T MRI pairs. The experimental results show that our framework provides more accurate and robust patient-specific STN localization than using state-of-the-art atlases. We also demonstrate the clinical feasibility of the proposed technique assessing the post-operative electrode active contact locations.

Optimization and comparative evaluation of nonlinear deformation algorithms for atlas-based segmentation of DBS target nuclei

Nonlinear registration of individual brain MRI scans to standard brain templates is common practice in neuroimaging and multiple registration algorithms have been developed and refined over the last 20 years. However, little has been done to quantitatively compare the available algorithms and much of that work has exclusively focused on cortical structures given their importance in the fMRI literature. In contrast, for clinical applications such as functional neurosurgery and deep brain stimulation (DBS), proper alignment of subcortical structures between template and individual space is important. This allows for atlas-based segmentations of anatomical DBS targets such as the subthalamic nucleus (STN) and internal pallidum (GPi). Here, we systematically evaluated the performance of six modern and established algorithms on subcortical normalization and segmentation results by calculating over 11,000 nonlinear warps in over 100 subjects. For each algorithm, we evaluated its performance using T1-or T2-weighted acquisitions alone or a combination of T1-, T2-and PD-weighted acquisitions in parallel. Furthermore, we present optimized parameters for the best performing algorithms. We tested each algorithm on two datasets, a state-of-the-art MRI cohort of young subjects and a cohort of subjects age- and MR-quality-matched to a typical DBS Parkinson's Disease cohort. Our final pipeline is able to segment DBS targets with precision comparable to manual expert segmentations in both cohorts. Although the present study focuses on the two prominent DBS targets, STN and GPi, these methods may extend to other small subcortical structures like thalamic nuclei or the nucleus accumbens.

Functional atlases for analysis of motor and neuropsychological outcomes after medial globus pallidus and subthalamic stimulation

Anatomical atlases have been developed to improve the targeting of basal ganglia in deep brain stimulation. However, the sole anatomy cannot predict the functional outcome of this surgery. Deep brain stimulation is often a compromise between several functional outcomes: motor, fluency and neuropsychological outcomes in particular. In this study, we have developed anatomo-clinical atlases for the targeting of subthalamic and medial globus pallidus deep brain stimulation. The activated electrode coordinates of 42 patients implanted in the subthalamic nucleus and 29 patients in the medial globus pallidus were studied. The atlas was built using the representation of the volume of tissue theoretically activated by the stimulation. The UPDRS score was used to represent the motor outcome. The Stroop test was represented as well as semantic and phonemic fluencies. For the subthalamic nucleus, best motor outcomes were obtained when the supero-lateral part of the nucleus was stimulated whereas the semantic fluency was impaired in this same region. For the medial globus pallidus, best outcomes were obtained when the postero ventral part of the nucleus was stimulated whereas the phonemic fluency was impaired in this same region. There was no significant neuropsychological impairment. We have proposed new anatomo-clinical atlases to visualize the motor and neuropsychological consequences at 6 months of subthalamic nucleus and pallidal stimulation in patients with Parkinson's disease.

Using automated electrode localization to guide stimulation management in DBS

Deep Brain Stimulation requires extensive postoperative testing of stimulation parameters to achieve optimal outcomes. Testing is typically not guided by neuroanatomical information on electrode contact locations. To address this, we present an automated reconstruction of electrode locations relative to the treatment target, the subthalamic nucleus, comparing different targeting methods: atlas‐, manual‐, or tractography‐based subthalamic nucleus segmentation. We found that most electrode contacts chosen to deliver stimulation were closest or second closest to the atlas‐based subthalamic nucleus target. We suggest that information on each electrode contact's location, which can be obtained using atlas‐based methods, might guide clinicians during postoperative stimulation testing.

Seed point discontinuity-based segmentation method for the substantia nigra and the red nucleus in quantitative susceptibility maps

BACKGROUND

The automatic segmentation of cerebral nuclei in the quantitative susceptibility mapping (QSM) images can provide assistance for surgical treatment and pathological mechanism studies. However, as the most frequently used segmentation method, the atlas method provides unsatisfactory results when segmenting the substantia nigra (SN) and the red nucleus (RN).

PURPOSE

To propose and evaluate an improved automatic method based on seed points-discontinuity for segmentations of the SN and the RN in QSM images.

STUDY TYPE

Prospective.

SUBJECTS

In all, 22 subjects, 11 patients with Parkinson's disease (PD), and 11 healthy subjects (mean age of 68.0 ± 6.9 years) underwent MR scans.

FIELD STRENGTH/SEQUENCE

3T system and a 3D multiecho gradient echo sequence with monopolar readout gradient.

ASSESSMENT

Manual segmentations by two radiologists (both with over 10 years of experience in neuroimaging) were used to establish a baseline for assessment. The Dice coefficient and the center-of-gravity distance was employed to evaluate the segmentation accuracy.

STATISTICAL TESTS

The mean value and standard deviation of the Dice coefficient and center-of-gravity distance were calculated separately to compare segmentation results from the proposed method, the level set method, the atlas method (including the single-atlas method and the multi-atlas majority voting method).

RESULTS

The statistical results of Dice coefficient of the SN and the RN between the ground truth and the segmentation were 0.79 ± 0.14 and 0.77 ± 0.06 for the proposed method, 0.40 ± 0.10 and 0.65 ± 0.09 for the level set method, 0.68 ± 0.09 and 0.64 ± 0.07 for the single-atlas method, 0.70 ± 0.06 and 0.68 ± 0.05 for the multi-atlas majority voting method, respectively. The proposed method also provides the lowest center-of-gravity distance value (1.05 ± 0.71 for the SN and 0.74 ± 0.35 for the RN).

DATA CONCLUSION

The segmentation results of the proposed method performed well on the in vivo data and were closer to the manual segmentation than the atlas method.

LEVEL OF EVIDENCE

1 Technical Efficacy: Stage 1 J. Magn. Reson. Imaging 2018;48:1112-1119.

Automatic preoperative planning of DBS electrode placement using anatomo-clinical atlases and volume of tissue activated

PURPOSE

Deep brain stimulation (DBS) is a procedure requiring accurate targeting and electrode placement. The two key elements for successful planning are preserving patient safety by ensuring a safe trajectory and creating treatment efficacy through optimal selection of the stimulation point. In this work, we present the first approach of computer-assisted preoperative DBS planning to automatically optimize both the safety of the electrode's trajectory and location of the stimulation point so as to provide the best clinical outcome.

METHODS

Building upon the findings of previous works focused on electrode trajectory, we added a set of constraints guiding the choice of stimulation point. These took into account retrospective data represented by anatomo-clinical atlases and intersections between the stimulation region and sensitive anatomical structures causing side effects. We implemented our method into automatic preoperative planning software to assess if the algorithm was able to simultaneously optimize electrode trajectory and the stimulation point.

RESULTS

Leave-one-out cross-validation on a dataset of 18 cases demonstrated an improvement in the expected outcome when using the new constraints. The distance to critical structures was not reduced. The intersection between the stimulation region and structures sensitive to stimulation was minimized.

CONCLUSIONS

Introducing these new constraints guided the planning to select locations showing a trend toward symptom improvement, while minimizing the risks of side effects, and there was no cost in terms of trajectory safety.

7T MRI subthalamic nucleus atlas for use with 3T MRI

Deep brain stimulation (DBS) of the subthalamic nucleus (STN) reduces motor symptoms in most patients with Parkinson disease (PD), yet may produce untoward effects. Investigation of DBS effects requires accurate localization of the STN, which can be difficult to identify on magnetic resonance images collected with clinically available 3T scanners. The goal of this study is to develop a high-quality STN atlas that can be applied to standard 3T images. We created a high-definition STN atlas derived from seven older participants imaged at 7T. This atlas was nonlinearly registered to a standard template representing 56 patients with PD imaged at 3T. This process required development of methodology for nonlinear multimodal image registration. We demonstrate mm-scale STN localization accuracy by comparison of our 3T atlas with a publicly available 7T atlas. We also demonstrate less agreement with an earlier histological atlas. STN localization error in the 56 patients imaged at 3T was less than 1 mm on average. Our methodology enables accurate STN localization in individuals imaged at 3T. The STN atlas and underlying 3T average template in MNI space are freely available to the research community. The image registration methodology developed in the course of this work may be generally applicable to other datasets.

Vocal emotion decoding in the subthalamic nucleus: An intracranial ERP study in Parkinson's disease

Using intracranial local field potential (LFP) recordings in patients with Parkinson's disease (PD) undergoing deep brain stimulation (DBS), we explored the electrophysiological activity of the subthalamic nucleus (STN) in response to emotional stimuli in the auditory modality. Previous studies focused on the influence of visual stimuli. To this end, we recorded LFPs within the STN in response to angry, happy, and neutral prosodies in 13 patients with PD who had just undergone implantation of DBS electrodes. We observed specific modulation of the right STN in response to anger and happiness, as opposed to neutral prosody, occurring at around 200-300ms post-onset, and later at around 850-950ms post-onset for anger and at around 3250-3350ms post-onset for happiness. Taken together with previous reports of modulated STN activity in response to emotional visual stimuli, the present results appear to confirm that the STN is involved in emotion processing irrespective of stimulus valence and sensory modality.

Improvement of Pyramidal Tract Side Effect Prediction Using a Data-Driven Method in Subthalamic Stimulation

OBJECTIVE

subthalamic nucleus deep brain stimulation (STN DBS) is limited by the occurrence of a pyramidal tract side effect (PTSE) induced by electrical activation of the pyramidal tract. Predictive models are needed to assist the surgeon during the electrode trajectory preplanning. The objective of the study was to compare two methods of PTSE prediction based on clinical assessment of PTSE induced by STN DBS in patients with Parkinson's disease.

METHODS

two clinicians assessed PTSE postoperatively in 20 patients implanted for at least three months in the STN. The resulting dataset of electroclinical tests was used to evaluate two methods of PTSE prediction. The first method was based on the volume of tissue activated (VTA) modeling and the second one was a data-driven-based method named Pyramidal tract side effect Model based on Artificial Neural network (PyMAN) developed in our laboratory. This method was based on the nonlinear correlation between the PTSE current threshold and the 3-D electrode coordinates. PTSE prediction from both methods was compared using Mann-Whitney U test.

RESULTS

1696 electroclinical tests were used to design and compare the two methods. Sensitivity, specificity, positive- and negative-predictive values were significantly higher with the PyMAN method than with the VTA-based method (P < 0.05).

CONCLUSION

the PyMAN method was more effective than the VTA-based method to predict PTSE.

SIGNIFICANCE

this data-driven tool could help the neurosurgeon in predicting adverse side effects induced by DBS during the electrode trajectory preplanning.

Image-guided preoperative prediction of pyramidal tract side effect in deep brain stimulation: proof of concept and application to the pyramidal tract side effect induced by pallidal stimulation

Deep brain stimulation of the medial globus pallidus (GPm) is a surgical procedure for treating patients suffering from Parkinson's disease. Its therapeutic effect may be limited by the presence of pyramidal tract side effect (PTSE). PTSE is a contraction time-locked to the stimulation when the current spreading reaches the motor fibers of the pyramidal tract within the internal capsule. The objective of the study was to propose a preoperative predictive model of PTSE. A machine learning-based method called PyMAN (PTSE model based on artificial neural network) accounting for the current used in stimulation, the three-dimensional electrode coordinates and the angle of the trajectory, was designed to predict the occurrence of PTSE. Ten patients implanted in the GPm have been tested by a clinician to create a labeled dataset of the stimulation parameters that trigger PTSE. The kappa index value between the data predicted by PyMAN and the labeled data was 0.78. Further evaluation studies are desirable to confirm whether PyMAN could be a reliable tool for assisting the surgeon to prevent PTSE during the preoperative planning.

Automated segmentation of the substantia nigra, subthalamic nucleus and red nucleus in 7T data at young and old age

With recent developments in MR acquisition at 7T, smaller brainstem structures such as the red nuclei, substantia nigra and subthalamic nuclei can be imaged with good contrast and resolution. These structures have important roles both in the study of the healthy brain and in diseases such as Parkinson's disease, but few methods have been described to automatically segment them. In this paper, we extend a method that we have previously proposed for segmentation of the striatum and globus pallidus to segment these deeper and smaller structures. We modify the method to allow more direct control over segmentation smoothness by using a Markov random field prior. We investigate segmentation performance in three age groups and show that the method produces consistent results that correspond well with manual segmentations. We perform a vertex-based analysis to identify changes with age in the shape of the structures and present results suggesting that the method may be at least as effective as manual delineation in capturing differences between subjects.

Magnetic Resonance Imaging Features of the Nigrostriatal System: Biomarkers of Parkinson's Disease Stages?

Introduction

Magnetic resonance imaging (MRI) can be used to identify biomarkers in Parkinson’s disease (PD); R2* values reflect iron content related to high levels of oxidative stress, whereas volume and/or shape changes reflect neuronal death. We sought to assess iron overload in the nigrostriatal system and characterize its relationship with focal and overall atrophy of the striatum in the pivotal stages of PD.

Methods

Twenty controls and 70 PD patients at different disease stages (untreated de novo patients, treated early-stage patients and advanced-stage patients with L-dopa-related motor complications) were included in the study. We determined the R2* values in the substantia nigra, putamen and caudate nucleus, together with striatal volume and shape analysis. We also measured R2* in an acute MPTP mouse model and in a longitudinal follow-up two years later in the early-stage PD patients.

Results

The R2* values in the substantia nigra, putamen and caudate nucleus were significantly higher in de novo PD patients than in controls. Early-stage patients displayed significantly higher R2* values in the substantia nigra (with changes in striatal shape), relative to de novo patients. Measurements after a two-year follow-up in early-stage patients and characterization of the acute MPTP mouse model confirmed that R2* changed rapidly with disease progression. Advanced-stage patients displayed significant atrophy of striatum, relative to earlier disease stages.

Conclusion

Each pivotal stage in PD appears to be characterized by putative nigrostriatal MRI biomarkers: iron overload at the de novo stage, striatal shape changes at early-stage disease and generalized striatal atrophy at advanced disease.

Boosting diagnosis accuracy of Alzheimer's disease using high dimensional recognition of longitudinal brain atrophy patterns

OBJECTIVE

Boosting accuracy in automatically discriminating patients with Alzheimer's disease (AD) and normal controls (NC), based on multidimensional classification of longitudinal whole brain atrophy rates and their intermediate counterparts in analyzing magnetic resonance images (MRI).

METHOD

Longitudinal percentage of brain volume changes (PBVC) in two-year follow up and its intermediate counterparts in early 6-month and late 18-month are used as features in supervised and unsupervised classification procedures based on K-mean, fuzzy clustering method (FCM) and support vector machine (SVM). The most relevant features for classification are selected using discriminative analysis (DA) of features and their principal components (PC). Accuracy of the proposed method is evaluated in a group of 30 patients with AD (16 males, 14 females, age±standard-deviation (SD)=75±1.36 years) and 30 normal controls (15 males, 15 females, age±SD=77±0.88 years) using leave-one-out cross-validation.

RESULTS

Results indicate superiority of supervised machine learning techniques over unsupervised ones in diagnosing AD and withal, predominance of RBF kernel over lineal one. Accuracies of 83.3%, 83.3%, 90% and 91.7% are achieved in classification by K-mean, FCM, linear SVM and SVM with radial based function (RBF) respectively.

CONCLUSION

Evidence that SVM classification of longitudinal atrophy rates may results in high accuracy is given. Additionally, it is realized that use of intermediate atrophy rates and their principal components improves diagnostic accuracy.

Segmentation of brain magnetic resonance images based on multi-atlas likelihood fusion: testing using data with a broad range of anatomical and photometric profiles

We propose a hierarchical pipeline for skull-stripping and segmentation of anatomical structures of interest from T1-weighted images of the human brain. The pipeline is constructed based on a two-level Bayesian parameter estimation algorithm called multi-atlas likelihood fusion (MALF). In MALF, estimation of the parameter of interest is performed via maximum a posteriori estimation using the expectation-maximization (EM) algorithm. The likelihoods of multiple atlases are fused in the E-step while the optimal estimator, a single maximizer of the fused likelihoods, is then obtained in the M-step. There are two stages in the proposed pipeline; first the input T1-weighted image is automatically skull-stripped via a fast MALF, then internal brain structures of interest are automatically extracted using a regular MALF. We assess the performance of each of the two modules in the pipeline based on two sets of images with markedly different anatomical and photometric contrasts; 3T MPRAGE scans of pediatric subjects with developmental disorders vs. 1.5T SPGR scans of elderly subjects with dementia. Evaluation is performed quantitatively using the Dice overlap as well as qualitatively via visual inspections. As a result, we demonstrate subject-level differences in the performance of the proposed pipeline, which may be accounted for by age, diagnosis, or the imaging parameters (particularly the field strength). For the subcortical and ventricular structures of the two datasets, the hierarchical pipeline is capable of producing automated segmentations with Dice overlaps ranging from 0.8 to 0.964 when compared with the gold standard. Comparisons with other representative segmentation algorithms are presented, relative to which the proposed hierarchical pipeline demonstrates comparative or superior accuracy.

Patch-based label fusion segmentation of brainstem structures with dual-contrast MRI for Parkinson's disease

PURPOSE

Parkinson's disease (PD) is a neurodegenerative disorder that impairs the motor functions. Both surgical treatment and study of PD require delineation of basal ganglia nuclei morphology. While many automatic volumetric segmentation methods have been proposed for the lentiform nucleus, few have attempted to identify the key brainstem substructures including the subthalamic nucleus (STN), substantia nigra (SN), and red nucleus (RN) due to their small size and poor contrast in conventional T1W MRI.

METHODS

A dual-contrast patch-based label fusion method was developed to segment the SN, STN, and RN using multivariate cross-correlation. Two different MRI contrasts (T2*w and phase) are produced from a multi-contrast multi-echo FLASH MRI sequence, enabling visualization of these nuclei. T1-T2* fusion MRI was used to resolve the issue of poor nuclei (i.e., the STN, SN, and RN) contrast on T1w MRI, and to mitigate susceptibility artifacts that may hinder accurate nonlinear registration on T2*w MRI. Unbiased group-wise registration was used for anatomical normalization between the atlas library and the target subject. The performance of the proposed method was compared with a state-of-the-art single-contrast label fusion technique.

RESULTS

The proposed method outperformed a state-of-the-art single-contrast patch-based method in segmenting the STN, RN and SN, and the results were better than those reported in previous literature.

CONCLUSION

Our dual-contrast patch-based label fusion method was superior to a single-contrast method for segmenting brainstem nuclei using a multi-contrast multi-echo FLASH MRI sequence. The method is promising for the treatment and research of Parkinson's disease. This method can be extended for multiple alternative image contrasts and other fields of applications.

Framework for integrated MRI average of the spinal cord white and gray matter: the MNI-Poly-AMU template

The field of spinal cord MRI is lacking a common template, as existing for the brain, which would allow extraction of multi-parametric data (diffusion-weighted, magnetization transfer, etc.) without user bias, thereby facilitating group analysis and multi-center studies. This paper describes a framework to produce an unbiased average anatomical template of the human spinal cord. The template was created by co-registering T2-weighted images (N = 16 healthy volunteers) using a series of pre-processing steps followed by non-linear registration. A white and gray matter probabilistic template was then merged to the average anatomical template, yielding the MNI-Poly-AMU template, which currently covers vertebral levels C1 to T6. New subjects can be registered to the template using a dedicated image processing pipeline. Validation was conducted on 16 additional subjects by comparing an automatic template-based segmentation and manual segmentation, yielding a median Dice coefficient of 0.89. The registration pipeline is rapid (~15 min), automatic after one C2/C3 landmark manual identification, and robust, thereby reducing subjective variability and bias associated with manual segmentation. The template can notably be used for measurements of spinal cord cross-sectional area, voxel-based morphometry, identification of anatomical features (e.g., vertebral levels, white and gray matter location) and unbiased extraction of multi-parametric data.

Investigation of morphometric variability of subthalamic nucleus, red nucleus, and substantia nigra in advanced Parkinson's disease patients using automatic segmentation and PCA-based analysis

Subthalamic nucleus (STN) deep brain stimulation (DBS) is an effective surgical therapy to treat Parkinson's disease (PD). Conventional methods employ standard atlas coordinates to target the STN, which, along with the adjacent red nucleus (RN) and substantia nigra (SN), are not well visualized on conventional T1w MRIs. However, the positions and sizes of the nuclei may be more variable than the standard atlas, thus making the pre-surgical plans inaccurate. We investigated the morphometric variability of the STN, RN and SN by using label-fusion segmentation results from 3T high resolution T2w MRIs of 33 advanced PD patients. In addition to comparing the size and position measurements of the cohort to the Talairach atlas, principal component analysis (PCA) was performed to acquire more intuitive and detailed perspectives of the measured variability. Lastly, the potential correlation between the variability shown by PCA results and the clinical scores was explored.

Analysis of electrode deformations in deep brain stimulation surgery

PURPOSE

Deep brain stimulation (DBS) surgery is used to reduce motor symptoms when movement disorders are refractory to medical treatment. Post-operative brain morphology can induce electrode deformations as the brain recovers from an intervention. The inverse brain shift has a direct impact on accuracy of the targeting stage, so analysis of electrode deformations is needed to predict final positions.

METHODS

DBS electrode curvature was evaluated in 76 adults with movement disorders who underwent bilateral stimulation, and the key variables that affect electrode deformations were identified. Non-linear modelling of the electrode axis was performed using post-operative computed tomography (CT) images. A mean curvature index was estimated for each patient electrode. Multivariate analysis was performed using a regression decision tree to create a hierarchy of predictive variables. The identification and classification of key variables that determine electrode curvature were validated with statistical analysis.

RESULTS

The principal variables affecting electrode deformations were found to be the date of the post-operative CT scan and the stimulation target location. The main pathology, patient's gender, and disease duration had a smaller although important impact on brain shift.

CONCLUSIONS

The principal determinants of electrode location accuracy during DBS procedures were identified and validated. These results may be useful for improved electrode targeting with the help of mathematical models.

Nonlocal regularization for active appearance model: Application to medial temporal lobe segmentation

The human medial temporal lobe (MTL) is an important part of the limbic system, and its substructures play key roles in learning, memory, and neurodegeneration. The MTL includes the hippocampus (HC), amygdala (AG), parahippocampal cortex (PHC), entorhinal cortex, and perirhinal cortex--structures that are complex in shape and have low between-structure intensity contrast, making them difficult to segment manually in magnetic resonance images. This article presents a new segmentation method that combines active appearance modeling and patch-based local refinement to automatically segment specific substructures of the MTL including HC, AG, PHC, and entorhinal/perirhinal cortex from MRI data. Appearance modeling, relying on eigen-decomposition to analyze statistical variations in image intensity and shape information in study population, is used to capture global shape characteristics of each structure of interest with a generative model. Patch-based local refinement, using nonlocal means to compare the image local intensity properties, is applied to locally refine the segmentation results along the structure borders to improve structure delimitation. In this manner, nonlocal regularization and global shape constraints could allow more accurate segmentations of structures. Validation experiments against manually defined labels demonstrate that this new segmentation method is computationally efficient, robust, and accurate. In a leave-one-out validation on 54 normal young adults, the method yielded a mean Dice κ of 0.87 for the HC, 0.81 for the AG, 0.73 for the anterior parts of the parahippocampal gyrus (entorhinal and perirhinal cortex), and 0.73 for the posterior parahippocampal gyrus.