Automatic localization of the subthalamic nucleus on patient-specific clinical MRI by incorporating 7 T MRI and machine learning: Application in deep brain stimulation.
ABSTRACT: 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.
Project description:OBJECTIVE:Deep brain stimulation (DBS) requires accurate localization of the anatomical target structure, and the precise placement of the DBS electrode within it. Ultra-high field 7 Tesla (T) MR images can be utilized to create patient-specific anatomical 3D models of the subthalamic nuclei (STN) to enhance pre-surgical DBS targeting as well as post-surgical visualization of the DBS lead position and orientation. We validated the accuracy of the 7T imaging-based patient-specific model of the STN and measured the variability of the location and dimensions across movement disorder patients. METHODS:72 patients who underwent DBS surgery were scanned preoperatively on 7T MRI. Segmentations and 3D volume rendering of the STN were generated for all patients. For 21 STN-DBS cases, microelectrode recording (MER) was used to validate the segmentation. For 12 cases, we computed the correlation between the overlap of the STN and volume of tissue activated (VTA) and the monopolar review for a further validation of the model's accuracy and its clinical relevancy. RESULTS:We successfully reconstructed and visualized the STN in all patients. Significant variability was found across individuals regarding the location of the STN center of mass as well as its volume, length, depth and width. Significant correlations were found between MER and the 7T imaging-based model of the STN (r = 0.86) and VTA-STN overlap and the monopolar review outcome (r = 0.61). CONCLUSION:The results suggest that an accurate visualization and localization of a patient-specific 3D model of the STN can be generated based on 7T MRI. The imaging-based 7T MRI STN model was validated using MER and patient's clinical outcomes. The significant variability observed in the STN location and shape based on a large number of patients emphasizes the importance of an accurate direct visualization of the STN for DBS targeting. An accurate STN localization can facilitate postoperative stimulation parameters for optimized patient outcome.
Project description:BACKGROUND:Implantation of deep brain stimulation (DBS) electrodes for the treatment of involuntary movement disorders, such as Parkinson's disease, routinely relies on the use of intraoperative electrophysiological confirmation to identify the optimal therapeutic target in the brain. However, only a few options exist to visualize the relative anatomic localization of intraoperative electrophysiological recordings with respect to post-operative imaging. We have developed a novel processing pipeline to visualize intraoperative electrophysiological signals registered to post-operative neuroanatomical imaging. NEW METHOD:We developed a processing pipeline built on the use of ITK-SNAP and custom MATLAB scripts to visualize the anatomical localization of intraoperative electrophysiological recordings mapped onto the post-operative MRI following implantation of DBS electrodes. This method combines the user-defined relevant electrophysiological parameters measured during the surgery with a manual segmentation of the DBS electrode from post-operative MRI; mapping the microelectrode recording (MER) depths along the DBS lead track. RESULTS:We demonstrate the use of our processing pipeline on data from Parkinson's disease patients undergoing DBS implantation targeted to the subthalamic nucleus (STN). The primary processing components of the pipeline are: extrapolation of the lead wire and alignment of intraoperative electrophysiology. CONCLUSION:We describe the use of a processing pipeline to aid clinicians and researchers engaged in deep brain stimulation work to correlate and visualize the intraoperative recording data with the post-operative DBS trajectory.
Project description:Awakening during deep brain stimulation (DBS) surgery may be stressful to patients. The aim of the current study was to evaluate the effect on MER signals and their applicability to subthalmic nucleus (STN) DBS surgery for patients with Parkinson's disease (PD) under sedation with propofol and fentanyl. Sixteen consecutive patients with PD underwent STN-DBS surgery with propofol and fentanyl. Their MER signals were achieved during the surgery. To identify the microelectrodes positions, the preoperative MRI and postoperative CT were used. Clinical profiles were also collected at the baseline and at 6 months after surgery. All the signals were slightly attenuated and contained only bursting patterns, compared with our previous report. All electrodes were mostly located in the middle one third part of the STN on both sides of the brain in the fused images. Six months later, the patients were improved significantly in the medication-off state and they met with less dyskinesia and less off-duration. Our study revealed that the sedation with propofol and fentanyl was applicable to STN-DBS surgery. There were no significant problems in precise positioning of bilateral electrodes. The surgery also improved significantly clinical outcomes in 6-month follow-up.
Project description:<h4>Background</h4>Subthalamic nucleus (STN) deep brain stimulation (DBS) represents a well-established treatment for patients with advanced Parkinson's disease (PD) insufficiently controlled with medical therapies. This study presents the long-term outcomes of patients with PD treated with STN-DBS using an MRI-guided/MRI-verified approach without microelectrode recording.<h4>Methods</h4>A cohort of 41 patients who underwent STN-DBS were followed for a minimum period of 5?years, with a subgroup of 12 patients being followed for 8-11?years. Motor status was evaluated using part III of the Unified Parkinson's Disease Rating Scale (UPDRS-III), in on- and off-medication/on-stimulation conditions. Preoperative and postoperative assessments further included activities of daily living (UPDRS-II), motor complications (UPDRS-IV), neuropsychological and speech assessments, as well as evaluation of quality of life. Active contacts localisation was calculated and compared with clinical outcomes.<h4>Results</h4>STN-DBS significantly improved the off-medication UPDRS-III scores, compared with baseline. However, UPDRS scores increased over time after DBS. Dyskinesias, motor fluctuations and demands in dopaminergic medication remained significantly reduced in the long term. Conversely, UPDRS-III on-medication scores deteriorated at 5 and 8?years, mostly driven by axial and bradykinesia subscores. Quality of life, as well as depression and anxiety scores, did not significantly change at long-term follow-up compared with baseline. In our series, severe cognitive decline was observed in 17.1% and 16.7% of the patients at 5 and 8?years respectively.<h4>Conclusions</h4>Our data confirm that STN-DBS, using an MRI-guided/MRI-verified technique, remains an effective treatment for motor 'off' symptoms of PD in the long term with low morbidity.
Project description:Background: The efficacy of deep brain stimulation (DBS) therapy in Parkinson's disease (PD) patients is highly dependent on the precise localization of the target structures such as subthalamic nucleus (STN). Most commonly, microelectrode single unit activity (SUA) recordings are performed to refine the target. This process is heavily experience based and can be technically challenging. Local field potentials (LFPs), representing the activity of a population of neurons, can be obtained from the same microelectrodes used for SUA recordings and allow flexible online processing with less computational complexity due to lower sampling rate requirements. Although LFPs have been shown to contain biomarkers capable of predicting patients' symptoms and differentiating various structures, their use in the localization of the STN in the clinical practice is not prevalent. Methods: Here we present, for the first time, a randomized and double-blinded pilot study with intraoperative online LFP processing in which we compare the clinical benefit from SUA- versus LFP-based implantation. Ten PD patients referred for bilateral STN-DBS were randomly implanted using either SUA or LFP guided targeting in each hemisphere. Although both SUA and LFP were recorded for each STN, the electrophysiologist was blinded to one at a time. Three months postoperatively, the patients were evaluated by a neurologist blinded to the intraoperative recordings to assess the performance of each modality. While SUA-based decisions relied on the visual and auditory inspection of the raw traces, LFP-based decisions were given through an online signal processing and machine learning pipeline. Results: We found a dramatic agreement between LFP- and SUA-based localization (16/20 STNs) providing adequate clinical improvement (51.8% decrease in 3-month contralateral motor assessment scores), with LFP-guided implantation resulting in greater average improvement in the discordant cases (74.9%, n = 3 STNs). The selected tracks were characterized by higher activity in beta (11-32 Hz) and high-frequency (200-400 Hz) bands (p < 0.01) of LFPs and stronger non-linear coupling between these bands (p < 0.05). Conclusion: Our pilot study shows equal or better clinical benefit with LFP-based targeting. Given the robustness of the electrode interface and lower computational cost, more centers can utilize LFP as a strategic feedback modality intraoperatively, in conjunction to the SUA-guided targeting.
Project description:Alongside stereotactic magnetic resonance imaging, microelectrode recording (MER) is frequently used during the deep brain stimulation (DBS) surgery for optimal target localization. The aim of this study is to optimize subthalamic nucleus (STN) mapping using MER analytical patterns. 16 patients underwent bilateral STN-DBS. MER was performed simultaneously for 5 microelectrodes in a setting of Ben's-gun pattern in awake patients. Using spikes and background activity several different parameters and their spectral estimates in various frequency bands including low frequency (2-7 Hz), Alpha (8-12 Hz), Beta (sub-divided as Low_Beta (13-20 Hz) and High_Beta (21-30 Hz)) and Gamma (31 to 49 Hz) were computed. The optimal STN lead placement with the most optimal clinical effect/side-effect ratio accorded to the maximum spike rate in 85% of the implantation. Mean amplitude of background activity in the low beta frequency range was corresponding to right depth in 85% and right location in 94% of the implantation respectively. MER can be used for STN mapping and intraoperative decisions for the implantation of DBS electrode leads with a high accuracy. Spiking and background activity in the beta range are the most promising independent parameters for the delimitation of the proper anatomical site.
Project description:Refinement of the subthalamic nucleus (STN) coordinates using intraoperative microelectrode recordings (MER) is routinely performed during deep brain stimulation (DBS) surgeries in Parkinson disease (PD). The commonly used criteria for electrophysiological localization of the STN are qualitative. The goal of this study was to validate quantitative STN detection algorithm (QD) derived from the multi-unit activity in a prospective setting.Ten PD patients underwent STN DBS surgery. The MUA was obtained by removing large spikes close to microelectrode using wavelet method and integrating the 500-2000Hz band in the power spectral density. The qualitative intraoperative mapping of the STN using MER (IOM) versus QD was compared using Bland-Altman and Pearson's correlation analysis.The clinical efficacy was confirmed in all subjects. The mean difference between IOM and QD of the dorsal/ventral border was 0.31±0.84/0.44±0.47mm. Using Bland-Altman statistic, only 2/36 (5.6%) differences (one for the dorsal border and one for the ventral border) were out of ±2 sd line of measurement differences. Correlation between dorsal border/ventral border positions obtained by IOM and QD was 0.79, p<0.0001/0.91, p<0.0001.Both methods are in reasonable agreement and are strongly correlated. The QD gives objective coordinates of the STN borders at high precision and may be more accurate than IOM. Prospective blinded comparative studies where the DBS leads will be placed using either QD or IOM are warranted.
Project description:<h4>Background</h4>Deep brain stimulation (DBS) of the subthalamic nucleus (STN) is an effective treatment for improving the motor symptoms of advanced Parkinson's disease (PD). Accurate positioning of the stimulation electrodes is necessary for better clinical outcomes.<h4>Objective</h4>We applied deep learning techniques to microelectrode recording (MER) signals to better predict motor function improvement, represented by the UPDRS part III scores, after bilateral STN DBS in patients with advanced PD. If we find the optimal stimulation point with MER by deep learning, we can improve the clinical outcome of STN DBS even under restrictions such as general anesthesia or non-cooperation of the patients.<h4>Methods</h4>In total, 696 4-second left-side MER segments from 34 patients with advanced PD who underwent bilateral STN DBS surgery under general anesthesia were included. We transformed the original signal into three wavelets of 1-50 Hz, 50-500 Hz, and 500-5,000 Hz. The wavelet-transformed MER was used for input data of the deep learning. The patients were divided into two groups, good response and moderate response groups, according to DBS on to off ratio of UPDRS part III score for the off-medication state, 6 months postoperatively. The ratio were used for output data in deep learning. The Visual Geometry Group (VGG)-16 model with a multitask learning algorithm was used to estimate the bilateral effect of DBS. Different ratios of the loss function in the task-specific layer were applied considering that DBS affects both sides differently.<h4>Results</h4>When we divided the MER signals according to the frequency, the maximal accuracy was higher in the 50-500 Hz group than in the 1-50 Hz and 500-5,000 Hz groups. In addition, when the multitask learning method was applied, the stability of the model was improved in comparison with single task learning. The maximal accuracy (80.21%) occurred when the right-to-left loss ratio was 5:1 or 6:1. The area under the curve (AUC) was 0.88 in the receiver operating characteristic (ROC) curve.<h4>Conclusion</h4>Clinical improvements in PD patients who underwent bilateral STN DBS could be predicted based on a multitask deep learning-based MER analysis.
Project description:Highlights • Neuroanatomical variations among patients are obscured in atlas-based VTA modeling.• N-of-1 neuroanatomical and VTA modeling enables patient-level precision.• Mean optimal stimulation is dorsomedial to the STN, near its posterior half.• Individual VTAs deviate from optimal stimulation sites to varying degrees.• Optimal stimulation sites for rigidity, bradykinesia, and tremor partially overlap. <h4>Background</h4> Motor outcomes after subthalamic deep brain stimulation (STN DBS) for Parkinson disease (PD) vary considerably among patients and strongly depend on stimulation location. The objective of this retrospective study was to map the regions of optimal STN DBS for PD using an atlas-independent, fully individualized (N-of-1) tissue activation modeling approach and to assess the relationship between patient-level therapeutic volumes of tissue activation (VTAs) and motor improvement. <h4>Methods</h4> The stimulation-induced electric field for 40 PD patients treated with bilateral STN DBS was modeled using finite element analysis. Neurostimulation models were generated for each patient, incorporating their individual STN anatomy, DBS lead position and orientation, anisotropic tissue conductivity, and clinical stimulation settings. A voxel-based analysis of the VTAs was then used to map the optimal location of stimulation. The amount of stimulation in specific regions relative to the STN was measured and compared between STNs with more and less optimal stimulation, as determined by their motor improvement scores and VTA. The relationship between VTA location and motor outcome was then assessed using correlation analysis. Patient variability in terms of STN anatomy, active contact position, and VTA location were also evaluated. Results from the N-of-1 model were compared to those from a simplified VTA model. <h4>Results</h4> Tissue activation modeling mapped the optimal location of stimulation to regions medial, posterior, and dorsal to the STN centroid. These regions extended beyond the STN boundary towards the caudal zona incerta (cZI). The location of the VTA and active contact position differed significantly between STNs with more and less optimal stimulation in the dorsal-ventral and anterior-posterior directions. Therapeutic stimulation spread noticeably more in the dorsal and posterior directions, providing additional evidence for cZI as an important DBS target. There were significant linear relationships between the amount of dorsal and posterior stimulation, as measured by the VTA, and motor improvement. These relationships were more robust than those between active contact position and motor improvement. There was high variability in STN anatomy, active contact position, and VTA location among patients. Spherical VTA modeling was unable to reproduce these results and tended to overestimate the size of the VTA. <h4>Conclusion</h4> Accurate characterization of the spread of stimulation is needed to optimize STN DBS for PD. High variability in neuroanatomy, stimulation location, and motor improvement among patients highlights the need for individualized modeling techniques. The atlas-independent, N-of-1 tissue activation modeling approach presented in this study can be used to develop and evaluate stimulation strategies to improve clinical outcomes on an individual basis.
Project description:OBJECTIVE:Intraoperative microelectrode recording (MER) and test-stimulation are regarded as the gold standard for proper placement of subthalamic (STN) deep brain stimulation (DBS) electrodes in Parkinson's disease (PD), requiring the patient to be awake during the procedure. In accordance with good clinical practice, most attending neurologists will request the clinically most efficacious trajectory for definite lead placement. However, the necessity of microelectrode-test-stimulation is disputed, as it may limit the access to DBS therapy, excluding those not willing or incapable of undergoing awake surgery. METHODS:We retrospectively analyzed the MERs and microelectrode-test-stimulation results with regard to the decision on definite lead placement and clinical outcome in a cohort of 67 PD-patients with STN-DBS. All patients received bilateral quadripolar ring electrodes. To ascertain overall procedural efficacy, we calculated the surgical index (SI) by comparing preoperative motor improvement induced by levodopa to that induced by stimulation 7 to 18 months after surgery, measured as the relative difference between ON and OFF-states on the Unified Parkinson's Disease Rating Scale motor part (UPDRS-3). Additionally, a side-specific surgical index (SSSI) was calculated using the unilateral assessable items of the UPDRS-3. The SSSI where microelectrode-test-stimulation overruled MER were compared to those where the result of microelectrode-test-stimulation was congruent to MER results. RESULTS:A total of 134 electrodes were analyzed. For final lead placement, the central trajectory was chosen in 54% of patient hemispheres. The mean SI was 0.99 (± 0.24). SSSI averaged 1.04 (± 0.45). In 37 lead placements, microelectrode-test-stimulation overruled MER in the final trajectory selection, in 27 of these lead placements adverse effects during microelectrode-test-stimulation were decisive. Neither the number of test electrodes used nor the STN-signal length had an impact on the SSSI. The SSSI did not differ between lead placements with MER/microelectrode-test-stimulation congruency and those where the results of microelectrode-test-stimulation initiated lead placement in a trajectory with shorter STN signal. CONCLUSION:Intraoperative testing is mandatory to ensure an optimal motor outcome of STN DBS in PD-patients when using quadripolar ring electrodes. However, we also demonstrated that neither the length of the STN-signal on MER nor the number of test electrodes influenced the motor outcome.