Deep Structure Research Articles

DGSIST: Clustering spatial transcriptome data based on deep graph structure Infomax

Although spatial transcriptomics data provide valuable insights into gene expression profiles and the spatial structure of tissues, most studies rely solely on gene expression information, underutilizing the spatial data. To fully leverage the potential of spatial transcriptomics and graph neural networks, the DGSI (Deep Graph Structure Infomax) model is proposed. This innovative graph data processing model uses graph convolutional neural networks and employs an unsupervised learning approach. It maximizes the mutual information between graph-level and node-level representations, emphasizing flexible sampling and aggregation of nodes and their neighbors. This effectively captures and incorporates local information from nodes into the overall graph structure. Additionally, this paper developed the DGSIST framework, an unsupervised cell clustering method that integrates the DGSI model, SVD dimensionality reduction algorithm, and k-means++ clustering algorithm. This aims to identify cell types accurately. DGSIST fully uses spatial transcriptomics data and outperforms existing methods in accuracy. Demonstrations of DGSIST’s capability across various tissue types and technological platforms have shown its effectiveness in accurately identifying spatial domains in multiple tissue sections. Compared to other spatial clustering methods, DGSIST excels in cell clustering and effectively eliminates batch effects without needing batch correction. DGSIST excels in spatial clustering analysis, spatial variation identification, and differential gene expression detection and directly applies to graph analysis tasks, such as node classification, link prediction, or graph clustering. Anticipation lies in the contribution of the DGSIST framework to a deeper understanding of the spatial organizational structures of diseases such as cancer.

Pallidal neuronal activity in Gilles de la Tourette syndrome and dystonic patients: A comparative study.

Gilles de la Tourette syndrome (GTS) and dystonia (DYS) are both hyperkinetic movement disorders effectively treated by deep brain stimulation (DBS) of the internal part of the globus pallidus (GPi). In this study, we compared single-neuron activity in the GPi between 18 GTS patients (with an average of 41 cells per patient) and 17 DYS patients (with an average of 54 cells per patient), all of whom underwent bilateral pallidal stimulation surgery, under general anesthesia or while awake at rest. We found no significant differences in GPi neuronal activity characteristics between patients operated on under general anesthesia versus those who were awake, irrespective of their diagnosis (GTS or DYS). We found higher firing rates, firing rate in bursts, pause duration and interspike interval coefficient of variation in GTS patients compared to DYS patients. On the opposite, we found higher number of pauses and bursts frequency in DYS patients. Lastly, we found a higher proportion of GPi oscillatory activities in DYS compared to GTS patients, with predominant activity within the low-frequency band (theta/alpha) in both patient groups. These findings underscore the complex relationship between the different neuronal discharge characteristic such as oscillatory or bursting activity within the GPi in shaping the clinical phenotypes of hyperkinetic disorders. Further research is warranted to deepen our understanding of how neuronal patterns are transmitted within deep brain structures and to develop strategies aimed at normalizing these pathological activities, by refining DBS techniques to enhance treatment efficacy and individual outcomes.

Determination of heat transfer characteristics of tube bundles over heating plate by machine learning

In this study machine learning is used with the aid of a deep neural network algorithm to predict convective heat transfer characteristics for inline and staggered tube bundles, and correlation equations for Nusselt number and friction factor are derived. Machine learning algorithm's data were obtained from experimental work for various transverse pitch of the tube bundles, longitudinal pitch of the tube bundles and Reynolds number. 276 experimental data points were taken for both inline and staggered tube bundles. However, considering that the data obtained from the experimental study may be insufficient for training, a two-step data augmentation method and retraining with cross-validation was used to prevent data deficiency in the deep neural network structure. Thus, the unseen data in the experimental work were also predicted. The coefficient of determination for the DNN model predictions was obtained greater than 0.96. One correlation equation for Nusselt Number and three correlation equations for friction factor were proposed from the augmented data with machine learning. The R2 values of the correlation equations varied between 89 % and 99 %. As a result, machine learning methods successfully applied to predict the Nusselt number and friction factor of tube bundles consistent with the experimental data.

Deep structure of the Kumkol basin in the northern Tibetan Plateau and its resource environmental implications

Deep structure of the Kumkol basin in the northern Tibetan Plateau and its resource environmental implications

Vehicle noise characteristics in magnetotelluric data and vehicle noise removal using waveform fitting

Magnetic field fluctuations due to vehicle noise were observed in magnetotelluric (MT) time-series data measured near roads. The observed vehicle noise had magnitudes ranging from tens to thousands of μA/m, whereas the observed weak natural MT signal magnitudes were approximately tens of μA/m. A small signal-to-noise ratio made it difficult to apply robust processing for removing vehicle noise. In addition, vehicle noise severely distorts the MT response in the MT deadband from 0.01 Hz to 0.3 Hz, where the MT signal is very weak, and methods to remove it are required for deep structure imaging. In this study, magnetic field fluctuations due to moving vehicles were simulated with a magnetic dipole and attempted to be removed using a waveform fitting method. A total of 378 vehicle noises were extracted from the near-road MT data and synthesized with the remote MT data without vehicle noises to investigate the effect of vehicle noise on the MT response. Removal of vehicle noise from synthesized remote MT data resulted in substantial restoration of the apparent resistivity and phase curves around the MT deadband and below 0.001 Hz. In the MT field data, the vehicle noise was simulated and removed with two moving dipoles; the magnitude of the remaining vehicle noise was reduced by approximately half compared to a single dipole, and very stable apparent resistivity and phase curves were obtained. Although electromagnetic noise distortion remains after vehicle noise removal, the waveform fitting method significantly improves the apparent resistivity and phase curve response in the 0.01–0.3 Hz frequency band.

Chaotic laser time series prediction based on an improved logistic mapping algorithm echo state network

Chaos synchronization plays vital functions in the fields of optical chaos secure communication. The synchronization performance can be significantly degraded by parameter mismatches between the chaotic transmitter and receiver. In this paper, the Deep-Logistical Mapping Echo State Network (D-LMESN) is proposed to enhance the performance of chaos synchronization. The network is upgraded by using an improved logical mapping algorithm and a deep reserve pool structure with phase space reconstruction. Results show that D-LMESN exhibits better performance in the prediction of chaotic time series, thanks to the adaptive parameter adjustment, which increases the ability to capture the dynamic characteristics of complex systems. Compared with ESN, the mean square error of this model is reduced by 55% and 72%, respectively, in chaotic laser simulation and actual data experiments. This provides a new possibility, to our knowledge, for the development of chaotic secure communication.

Enhancing Clustering Performance of Failed Test Cases during HIL Simulation: A Study on Deep Auto-Encoder Structures and Hyperparameter Tuning

Over the last decade, hardware-in-the-loop (HIL) simulation has been established as a safe, efficient, reliable, and flexible method for performing real-time simulation. Furthermore, in the automotive sector, the HIL system has been recommended in the ISO 26262 standard as a powerful platform for performing realistic simulation during system integration testing. As a result of performing HIL black-box tests, the results of executing test cases (TCs) are reported as pass/fail without determining the nature and root causes of the underlying failures. The conventional analysis process of the failed TCs relies on expert knowledge. The higher the number of failed TCs, the higher the cost of manual analysis in terms of time and effort. In light of the shortcomings of existing methodologies, this study presents a novel intelligent framework capable of analyzing failed TCs without the need for expert knowledge or code access. To this end, a convolutional auto-encoder-based deep-learning approach is proposed to extract representative features from the textual description of the failed TCs. Furthermore, k-means-based clustering is used to categorize the extracted features according to their respective failure classes. To illustrate the effectiveness and validate the performance of the proposed method, a virtual test drive with real-time HIL simulation is presented as a case study. The proposed model exhibits superior clustering performance compared to other standalone k-means algorithms, as demonstrated by the David Bouldin Index (DBI) and accuracy values of 0.5184 and 94.33%, respectively. Furthermore, the proposed model shows a significant advantage in terms of feature extraction and clustering performance compared to the current state-of-the-art fault-analysis method. The proposed approach not only supports the validation process and improves the safety and reliability of the systems but also reduces the costs of manual analysis in terms of time and effort.

Spinal Diffuse Midline Glioma H3 K27M-Altered: Report of a Rare Tumor with Extracranial Skeletal Metastases and Review of Literature.

Diffuse midline glioma, H3 K27-altered is a rare and aggressive pediatric brain tumor with a grim prognosis. Diffuse midline glioma is characterized by specific molecular alterations, including H3 K27 mutations, and involves deep midline structures such as the brainstem, cerebellum, spinal cord, and thalamus. These tumors present with a classic triad of symptoms and have limited surgical options due to their challenging locations. Extra-neural metastases are an unusual occurrence in diffuse midline glioma and have been rarely described. Here we report a 17-year-old girl with spinal diffuse midline glioma, H3 K27M-mutant, who presented with multiple metastatic osseous lesions confirmed on biopsy of the thoracic vertebral lesion. Due to the rapid disease progression, the patient was recommended palliative therapy. Extra-neural metastases in diffuse midline glioma are rare, with only 16 reported patients, and no standard therapy exists. An accurate and early diagnosis is necessary to develop a personalized plan of treatment. Further research is needed to gain insights into the molecular pathology of diffuse midline glioma, H3 K27-altered, and improve the quality of life and the outcome of patients with this deadly disease.

Seismic imaging of deep‐seated gold deposit and host rocks through a reappraisal of legacy seismic data in the Fochville mining area, South Africa

AbstractReappraisal of legacy reflection seismic data has shown to deliver value in mineral exploration, particularly in brownfield settings. In this work, we demonstrate how the reappraisal and processing of legacy reflection seismic data can be advantageous in the mineral exploration industry. We use today's standard seismic processing tools to improve the imaging of deep and complex geological structures that host mineral deposits. The recovered and processed 25.3 km long legacy seismic profile in this study was acquired in 1983 by the Gold Division of Anglo‐American as part of the Witwatersrand Gold Fields exploration program. This study aims to improve the imaging of the Ventersdorp Contact Reef gold‐bearing horizon (termed reef), a world‐class gold deposit (2 m thick) situated at depths between ∼2400 and ∼4100 m below the ground surface near South Deep Gold Mine in Fochville, South Africa. The final processing results from the pre‐stack time and phase‐shift migration approaches clearly reveal a dipping reflection associated with the gold‐bearing horizon and major steeply dipping faults that crosscut and displace the deposit. The final results are integrated with borehole information, 1D synthetic modelling and aeromagnetic data to constrain the structural interpretation. In particular, 1D synthetic simulation and borehole data constrain the depth position of the gold deposit. The magnetic data provides additional constraints on the complex faulted blocks of the host rocks such as the intrusions that may have a direct impact on ore resources and evaluation. The mining companies, such as South Deep Gold Mine, operating closer to the seismic profiles can use this new structural information to update the current geological models and improve future mine planning and designs, thus providing some insight into the prospectivity of unmined ground.

Arc-like curvilinear bullet pathways along inner surface of skull - two cases.

To provide example cases within the peer-reviewed literature of a phenomenon related to certain gunshot wounds of the head where the bullet pathway courses along the inner table of the skull. Presentation of two cases discovered at medicolegal autopsy. The two cases provide radiologic and autopsy demonstration of bullet pathways within the head where the projectiles track along the inner aspect of the cranium, in a circumferential manner, such that the bullets are recovered in a location on the opposite side of the brain/head, with no projectile pathway within the intervening deep structures of the brain. The cases provide photographic and radiologic autopsy documentation of the phenomenon, which is described in forensic pathology textbooks and rare peer-reviewed literature sources, but without photographic demonstration. Recognition of this phenomenon in the clinical setting, or when correlating autopsy external and radiologic findings with historical information, may provide an explanation for situations where immediate incapacitation does not occur following a gunshot wound of the head.

Mechanical Characteristics of Deep Excavation Support Structure with Asymmetric Load on Ground Surface

The purpose of this paper is to capture the mechanical response of the support structure of deep excavation subject asymmetric load. A two-dimensional (2D) numerical analysis model was established by taking a pipe gallery deep excavation subject to asymmetric load as an example. The numerical analysis results were in good agreement with the measured data, thus verified the validity of the numerical model. On this basis, the stress and displacement of support structure caused by the change in foundation asymmetric load were studied. According to the numerical results, horizontal displacement of the diaphragm wall (DW) was dominant, and the maximum horizontal displacement of the DW was 7.54 mm when the deep excavation was completed. With the increase in asymmetric load, the left wall displacement continued to increase, while the displacement of the right DW continued to decrease, and the maximum horizontal wall displacement occurred near the excavation face. The DW was the main bending component, and the maximum wall bending moment when the deep excavation was completed was 173.5 kN·m. The maximum wall bending moment increased with the increase in asymmetric load, and the maximum wall bending moment on the left of the deep excavation was greater than that on the right. The inner support sustained the main component of axial force, with the axial force peaking at 1051.8 kN when the deep excavation was completed. The axial force of the inner support increased with increasing the asymmetric load, and the axial force of the second inner support was obviously greater than that of the first inner support. This research has a positive effect on the design and optimization of deep excavation support structure subject to asymmetric load on ground surface.

Generalizing deep learning electronic structure calculation to the plane-wave basis.

Deep neural networks capable of representing the density functional theory (DFT) Hamiltonian as a function of material structure hold great promise for revolutionizing future electronic structure calculations. However, a notable limitation of previous neural networks is their compatibility solely with the atomic-orbital (AO) basis, excluding the widely used plane-wave (PW) basis. Here we overcome this critical limitation by proposing an accurate and efficient real-space reconstruction method for directly computing AO Hamiltonian matrices from PW DFT results. The reconstruction method is orders of magnitude faster than traditional projection-based methods to convert PW results to the AO basis, and the reconstructed Hamiltonian matrices can faithfully reproduce the PW electronic structure, thus bridging the longstanding gap between the AO basis deep learning electronic structure approach and PW DFT. Advantages of the PW methods, such as high accuracy, high flexibility and wide applicability, thus can be all integrated into deep learning electronic structure methods without sacrificing these methods' inherent benefits. This allows for the construction of large-scale and high-fidelity training datasets with the help of PW DFT results towards the development of precise and broadly applicable deep learning electronic structure models.

NAS FD Lung: A novel lung assist diagnostic system based on neural architecture search

In the detection and recognition of lung nodules, pulmonary nodules vary in size and shape and contain many similar tissues and organs around them, leading to the problems of both missed detection and false detection in existing detection algorithms. Designing proprietary detection and recognition networks manually requires substantial professional expertise. This process is time-consuming and labour-intensive and leads to issues like parameter redundancy and improper feature selection. Therefore, this paper proposes a new pulmonary CAD (computer-aided diagnosis) system for pulmonary nodules, NAS FD Lung (Using the NAS approach to search deep FPN and DPN networks), that can automatically learn and generate a deep learning network tailored to pulmonary nodule detection and recognition task requirements. NAS FD Lung aims to use automatic search to generate deep learning networks in the auxiliary diagnosis of pulmonary nodules to replace the manual design of deep learning networks. NAS FD Lung comprises two automatic search networks: BM NAS-FPN (Using NAS methods to search for deep FPN structures with Binary operation and Matrix multiplication fusion methods) network for nodule detection and NAS-A-DPN (Using the NAS approach to search deep DPN networks with attention mechanism) for nodule identification. The proposed technique is tested on the LUNA16 dataset, and the experimental results show that the model is superior to many existing state-of-the-art approaches. The detection accuracy of lung nodules is 98.23%. Regarding the lung nodules classification, the accuracy, specificity, sensitivity, and AUC values achieved 96.32%,97.14%,95.82%, and 98.33%, respectively.

Industrial Process Fault Detection Based on Siamese Recurrent Autoencoder

Although deep autoencoders excel at extracting intricate features, their application in process monitoring is limited by the requirement for large sample sizes and interpretability of latent representations. This work presents a special deep learning structure named Siamese network to detect abnormal deviations in nonlinear dynamic processes. By leveraging the capability of Siamese architecture to process multiple inputs simultaneously, the training sample size expands exponentially, which enhances the learning potential of the model. Furthermore, a long short-term memory unit is integrated to enable the capture of long-term process dynamics. To refine the distribution of latent features extracted from diverse data types, a contrastive loss function is proposed, which strengthens the model's fault detection capabilities and enhances its interpretation of latent representations. Then T2 statistic is established on the latent space to perform fault detection. The effectiveness of the method is demonstrated through case studies on simulation processes and an industrial process.

Biochars from waste as an additive material in anaerobic co-digestion: Characterization and application in batch and semi-continuous systems

The conversion of organic waste into energy through anaerobic digestion (AD) is a profitable and environmentally friendly strategy. Enhancing the process is crucial for optimizing and maximizing biogas and methane production. To achieve this, various strategies can be employed, such as anaerobic co-digestion (co-AD) and the use of biochar. Biochar aids in process stability, provides buffering capacity, mitigates inhibitory toxins, immobilizes microorganisms, facilitates microbial colonization, and accelerates electron transfer between microorganisms. This study aimed to produce biochar from different wastes and pyrolysis temperatures and to investigate its use as an additive material in the co-AD of fruit and vegetable waste (FVW) and laying chicken manure (LCM) in batch and semi-continuous systems. Thermogravimetric analysis (TGA) and differential scanning calorimetry were performed to determine the pyrolysis temperature for biochar production, defining temperatures of 450 and 550 °C. A biochemical methane potential (BMP) test was conducted in 120 mL reactors (working volume) with the addition of 10 g/L of biochar from FVW, LCM, and wood pruning waste (WPW). WPW450 biochar showed the highest fixed carbon content, rough surface, and deep porous structure, factors that contributed to a 28 % increase in methane production compared to the control. Semi-continuous tests were conducted in 50 L reactors (working volume) using a reduced dosage (1 g/L) of WPW450 biochar, providing process stability and better biogas quality, resulting in a 31 % increase in methane production. The reduction in biochar dosage did not negatively impact the semi-continuous system, which showed a 74 % higher methane yield compared to the batch system. The results of this study highlight the importance of biochar addition to increase methane production in the co-AD of FVW and LCM at different scales, as well as the study of different biomasses and temperatures for biochar production.

Deep structural insights into the origin of the Heyuan ms 4.5 revealed by deep seismic sounding profiles in southeast China

This study presents an interpretation of a deep seismic sounding (DSS) profile that carried out along the Cathaysia Block in southeast china, aiming to explore the crustal velocity structure. Data used in the survey were obtained from three controlled-source explosions conducted along the 320 km long Lianping-Heyuan-Shanwei profile. The modeling was based on ray tracing, using the extrapolation of seismic wave arrival times with the help of travel times predicted from a one-dimensional velocity model. The average velocity structure of the middle crust is 6.0–6.4 km/s, while a low velocity anomaly of approximately 0.1–0.2 km/s in the vicinity of the Heyuan-Shaowu fault zone. The resulting 2D velocity model indicates that steeply dipping low-velocity zones that correlate with the projection of two major fault zones. These zones, together with a flat LVZ at a depth of 12 km, define a triangular region that correlates with numerous hypocenters. This tectonic setting is favorable for the accumulation and release of strain in high-velocity media within the triangular region. The unique triangular structure in the upper crust provides necessary shallow medium conditions for seismic activity. This indicates that increased seismicity within this area is partially attributed to heightened stress within higher-velocity material. The triangular annular low-velocity body, situated in the upper crust, is influenced by dynamic environmental factors caused by deep thermal disturbances. The deep-seated fault serves as a conduit for the historical migration of thermal material, likely contributing to the seismogenic conditions for earthquakes in Heyuan’s region through deep-seated thermal disturbances. These findings provide a novel geophysical reference model for the regional seismicity near the Xinfengjang reservoir and significantly contribute to understanding the causal relationship between tectonic setting and seismicity. In comparison with previous studies, our research is dedicated to investigating the causes of shallow earthquakes in the region and exploring the relationship between deep and shallow structures.

Lingual Actinomycosis – A Rare Report of the Ray Fungus

Actinomycosis is an uncommon chronic suppurative infection caused by a Gram-positive anerobic bacterium Actinomyces israelii or Actinomyces gerencseriae. Lingual actinomycosis is a rare form of actinomycosis with a prevalence of 3%. We report a case of a 16-year-old female, with an acute-onset painful ulcer with a yellowish-white slough over the left lateral border of the tongue preceded by a tongue swelling. There was a history of tongue bites before the onset of swelling. Histopathological examination was suggestive of lingual actinomycosis. The lesions resolved 3 weeks after treatment with amoxicillin and clavulanic acid combination, but the patient was advised to complete the 6-week antibiotic course. The patient did not have any new ulcers in the follow-up. The tongue is inherently resistant to bacterial adherence and infection, but a break in the integrity of the skin or the presence of devitalized tissue usually aids in the invasion of the organism into deeper structures. The disease is usually localized to the tongue but can also spread to jaws (osteomyelitis of jaws), lungs, heart, and brain and can lead to death. Although antibiotic therapy remains the mainstay of treatment, surgical excision has a profound positive effect on the antimicrobial treatment. It is worth considering actinomycosis as a possibility in suppurative, granulomatous, and suspicious lesions of the tongue.

State of the Art of Brain Function Detection Technologies in Robot-Assisted Lower Limb Rehabilitation.

Background: With an aging population, the prevalence of neurological disorders is increasing, leading to a rise in lower limb movement disorders and, in turn, a growing need for rehabilitation training. Previous neuroimaging studies have shown a growing scientific interest in the study of brain mechanisms in robot-assisted lower limb rehabilitation (RALLR). Objective: This review aimed to determine differences in neural activity patterns during different RALLR tasks and the impact on neurofunctional plasticity. Methods: Sixty-five articles in the field of RALLR were selected and tested using three brain function detection technologies. Results: Most studies have focused on changes in activity in various regions of the cerebral cortex during different lower limb rehabilitation tasks but have also increasingly focused on functional changes in other cortical and deep subcortical structures. Our analysis also revealed a neglect of certain task types. Conclusion: We identify and discuss future research directions that may contribute to a clear understanding of neural functional plasticity under different RALLR tasks. Impact Statement The evaluation of robot-assisted lower limb rehabilitation based on brain function detection technology can assess the neurological changes of patients in the rehabilitation process by monitoring brain activities and can also provide more accurate guidance for robot-assisted lower limb rehabilitation. By monitoring the patient's brain activity, the robot can adjust according to the real-time status of the patient to achieve more effective rehabilitation training. This has potential impact on improving the rehabilitation effect and speeding up the rehabilitation process of patients.

CyberAIBot: Artificial Intelligence in an Intrusion Detection System for CyberSecurity in the IoT

The cyber ecosystem presents two interesting properties. Attackers and defenders are normally the same entity; a detailed knowledge of defensive strategies optimises an attack whereas a good defence, based on a layered structure also includes attacks. In addition, Artificial Intelligence (AI) provides new techniques and tools to both attackers and defenders such as Generative Adversarial Networks (GANs) based on Generators and Discriminators for impersonation or Deep Learning (DL) for exhaustive scanning. To address this dilemma, this article presents CyberAIBot: Artificial Intelligence in an Intrusion Detection System for CyberSecurity in the Internet of Things (IoT) aimed at Operational Technology (OT) and Information Technology (IT) network traffic. CyberAIBot is based on a Deep Learning management structure in a private or local edge cloud computing approach where AI makes decisions as close as possible to the source of data. CyberAIBot gradually detects, learns, and adapts to different cyber attacks. In detail, CyberAIBot uses Deep Learning (DL) technical clusters trained in specific attacks and a Deep Learning (DL) management cluster specialises in taking management decisions rather than technical evaluations. This management cluster supervises conflicting classifications from the deep learning technical clusters. CyberAIbot is trained against several datasets and its performance is evaluated between two classification algorithms, the Long Short-Term Memory (LSTM) networks and Support Vector Machines (SVMs). The SVM DL clusters learn faster (average 15x) although the LSTM DL clusters perform better (average 30%). The LSTM DL management cluster performs better than the SVM DL management cluster although it recognises fewer traffic types. The total number of data points analysed by CyberAIBot is 5.52E+08, equivalent to the distance between the planet Earth to the Moon in meters.

Brain metabolic response to repetitive transcranial magnetic stimulation to lesion network in cervical dystonia

BackgroundA previous study identified a brain network underlying cervical dystonia (CD) based on causal brain lesions. This network was shown to be abnormal in idiopathic CD and aligned with connections mediating treatment response to deep brain stimulation, suggesting generalizability across etiologies and relevance for treatment. The main nodes of this network were located in the deep cerebellar structures and somatosensory cortex (S1), the latter of which can be easily reached via non-invasive brain stimulation. To date, there are no studies testing brain stimulation to networks identified using lesion network mapping. ObjectivesTo assess target engagement by stimulating the S1 and testing the brain's acute metabolic response to repetitive transcranial magnetic stimulation in CD patients and healthy controls. MethodsThirteen CD patients and 14 controls received a single session of continuous theta burst (cTBS) and sham to the right S1. Changes in regional brain glucose metabolism were measured using [18F]FDG-PET. ResultscTBS increased metabolism at the stimulation site in CD (P = 0.03) but not in controls (P = 0.15; group difference P = 0.01). In subcortical regions, cTBS increased metabolism in the brainstem in CD only (PFDR = 0.04). The remote activation was positively associated with dystonia severity and efficacy of sensory trick phenomenon in CD patients. ConclusionsOur results provide further evidence of abnormal sensory system function in CD and show that a single session of S1 cTBS is sufficient to induce measurable changes in brain glucose metabolism. These findings support target engagement, motivating therapeutic trials of cTBS to the S1 in CD.