Parkinsonian tremor identification with multiple local field potential feature classification

Parkinsonian rest tremor can be distinguished from voluntary hand movements based on subthalamic and cortical activity

OBJECTIVE

To distinguish Parkinsonian rest tremor and different voluntary hand movements by analyzing brain activity.

METHODS

We re-analyzed magnetoencephalography and local field potential recordings from the subthalamic nucleus of six patients with Parkinson's disease. Data were obtained after withdrawal from dopaminergic medication (Med Off) and after administration of levodopa (Med On). Using gradient-boosted tree learning, we classified epochs as tremor, fist-clenching, forearm extension or tremor-free rest.

RESULTS

Subthalamic activity alone was insufficient for distinguishing the four different motor states (balanced accuracy mean: 38%, std: 7%). The combination of cortical and subthalamic features, in contrast, allowed for a much more accurate classification (balanced accuracy mean: 75%, std: 17%). Adding a single cortical area improved balanced accuracy by 17% on average, as compared to classification based on subthalamic activity alone. In most patients, the most informative cortical areas were sensorimotor cortical regions. Decoding performance was similar in Med On and Med Off.

CONCLUSIONS

Electrophysiological recordings allow for distinguishing several motor states, provided that cortical signals are monitored in addition to subthalamic activity.

SIGNIFICANCE

By combining cortical recordings, subcortical recordings and machine learning, adaptive deep brain stimulation systems might be able to detect tremor specifically and to respond adequately to several motor states.

Parkinson's disease tremor prediction using EEG data analysis-A preliminary and feasibility study

PURPOSE

Tremor is one of the hallmarks of Parkinson's disease (PD) that does not respond effectively to conventional medications. In this regard, as a complementary solution, methods such as deep brain stimulation have been proposed. To apply the intervention with minimal side effects, it is necessary to predict tremor initiation. The purpose of the current study was to propose a novel methodology for predicting resting tremors using analysis of EEG time-series.

METHODS

A modified algorithm for tremor onset detection from accelerometer data was proposed. Furthermore, a machine learning methodology for predicting PD hand tremors from EEG time-series was proposed. The most discriminative features extracted from EEG data based on statistical analyses and post-hoc tests were used to train the classifier for distinguishing pre-tremor conditions.

RESULTS

Statistical analyses with post-hoc tests showed that features such as form factor and statistical features were the most discriminative features. Furthermore, limited numbers of EEG channels (F3, F7, P4, CP2, FC6, and C4) and EEG bands (Delta and Gamma) were sufficient for an accurate tremor prediction based on EEG data. Based on the selected feature set, a KNN classifier obtained the best pre-tremor prediction performance with an accuracy of 73.67%.

CONCLUSION

This feasibility study was the first attempt to show the predicting ability of EEG time-series for PD hand tremor prediction. Considering the limitations of this study, future research with longer data, and different brain dynamics are needed for clinical applications.

Hyper-parameter tuning and feature extraction for asynchronous action detection from sub-thalamic nucleus local field potentials

Introduction

Decoding brain states from subcortical local field potentials (LFPs) indicative of activities such as voluntary movement, tremor, or sleep stages, holds significant potential in treating neurodegenerative disorders and offers new paradigms in brain-computer interface (BCI). Identified states can serve as control signals in coupled human-machine systems, e.g., to regulate deep brain stimulation (DBS) therapy or control prosthetic limbs. However, the behavior, performance, and efficiency of LFP decoders depend on an array of design and calibration settings encapsulated into a single set of hyper-parameters. Although methods exist to tune hyper-parameters automatically, decoders are typically found through exhaustive trial-and-error, manual search, and intuitive experience.

Methods

This study introduces a Bayesian optimization (BO) approach to hyper-parameter tuning, applicable through feature extraction, channel selection, classification, and stage transition stages of the entire decoding pipeline. The optimization method is compared with five real-time feature extraction methods paired with four classifiers to decode voluntary movement asynchronously based on LFPs recorded with DBS electrodes implanted in the subthalamic nucleus of Parkinson's disease patients.

Results

Detection performance, measured as the geometric mean between classifier specificity and sensitivity, is automatically optimized. BO demonstrates improved decoding performance from initial parameter setting across all methods. The best decoders achieve a maximum performance of 0.74 ± 0.06 (mean ± SD across all participants) sensitivity-specificity geometric mean. In addition, parameter relevance is determined using the BO surrogate models.

Discussion

Hyper-parameters tend to be sub-optimally fixed across different users rather than individually adjusted or even specifically set for a decoding task. The relevance of each parameter to the optimization problem and comparisons between algorithms can also be difficult to track with the evolution of the decoding problem. We believe that the proposed decoding pipeline and BO approach is a promising solution to such challenges surrounding hyper-parameter tuning and that the study's findings can inform future design iterations of neural decoders for adaptive DBS and BCI.

Diagnosis and Treatment of Tremor in Parkinson's Disease Using Mechanical Devices

BACKGROUND

Parkinsonian tremors are sometimes confused with essential tremors or other conditions. Recently, researchers conducted several studies on tremor evaluation using wearable sensors and devices, which may support accurate diagnosis. Mechanical devices are also commonly used to treat tremors and have been actively researched and developed. Here, we aimed to review recent progress and the efficacy of the devices related to Parkinsonian tremors.

METHODS

The PubMed and Scopus databases were searched for articles. We searched for "Parkinson disease" and "tremor" and "device".

RESULTS

Eighty-six articles were selected by our systematic approach. Many studies demonstrated that the diagnosis and evaluation of tremors in patients with PD can be done accurately by machine learning algorithms. Mechanical devices for tremor suppression include deep brain stimulation (DBS), electrical muscle stimulation, and orthosis. In recent years, adaptive DBS and optimization of stimulation parameters have been studied to further improve treatment efficacy.

CONCLUSIONS

Due to developments using state-of-the-art techniques, effectiveness in diagnosing and evaluating tremor and suppressing it using these devices is satisfactorily high in many studies. However, other than DBS, no devices are in practical use. To acquire high-level evidence, large-scale studies and randomized controlled trials are needed for these devices.

A Kinematic Data-Driven Approach to Differentiate Involuntary Choreic Movements in Individuals With Neurological Conditions

OBJECTIVE

The ability to differentiate similar choreic involuntary movements could lay the groundwork for the development of a minimally-invasive screening tool for their etiology and provide in-depth understandings of pathophysiology. As a first step, we investigate kinematic differences between Huntington's disease (HD) chorea and Parkinson's disease (PD) choreic levodopa-induced dyskinesia (LID), which have distinct pathological causes yet share a great kinematic resemblance.

METHODS

Twenty subjects with HD and ten subjects with PD stood with both upper limbs in front of them for approximately 60 seconds. The three-dimensional velocity time-series of involuntary movements of both hands were segmented into one-dimensional sub-movements abutted by velocity zero-crossings. A combination of unsupervised and supervised machine learning algorithms was employed to automatically select data features extracted from sub-movements and distinguish the two types of involuntary choreic movements.

RESULTS

The trained model was able to accurately classify chorea vs. LID with an Area Under the Receiver Operating Characteristic Curve of 99.5%. A set of important features contributing to the construction of the classification model were identified and investigated.

CONCLUSION

The trained model may serve as a tool for the automatic identification of different types of involuntary choreic movements, enabling continuous monitoring and personalized treatment for patients in various clinical settings.

SIGNIFICANCE

The results provide insights into kinematic characteristics of HD chorea and PD LID, which is the first step towards an improved general understanding of involuntary choreic movements.

Machine learning based brain signal decoding for intelligent adaptive deep brain stimulation

Sensing enabled implantable devices and next-generation neurotechnology allow real-time adjustments of invasive neuromodulation. The identification of symptom and disease-specific biomarkers in invasive brain signal recordings has inspired the idea of demand dependent adaptive deep brain stimulation (aDBS). Expanding the clinical utility of aDBS with machine learning may hold the potential for the next breakthrough in the therapeutic success of clinical brain computer interfaces. To this end, sophisticated machine learning algorithms optimized for decoding of brain states from neural time-series must be developed. To support this venture, this review summarizes the current state of machine learning studies for invasive neurophysiology. After a brief introduction to the machine learning terminology, the transformation of brain recordings into meaningful features for decoding of symptoms and behavior is described. Commonly used machine learning models are explained and analyzed from the perspective of utility for aDBS. This is followed by a critical review on good practices for training and testing to ensure conceptual and practical generalizability for real-time adaptation in clinical settings. Finally, first studies combining machine learning with aDBS are highlighted. This review takes a glimpse into the promising future of intelligent adaptive DBS (iDBS) and concludes by identifying four key ingredients on the road for successful clinical adoption: i) multidisciplinary research teams, ii) publicly available datasets, iii) open-source algorithmic solutions and iv) strong world-wide research collaborations.

Neurophysiology during movement disorder surgery

During stereotactic procedures for treating medically refractory movement disorders, intraoperative neurophysiology shifts its focus from simply monitoring the effects of surgery to an integral part of the surgical procedure. The small size, poor visualization, and physiologic nature of these deep brain targets compel the surgeon to rely on some form of physiologic for confirmation of proper anatomic targeting. Even given the newer reliance on imaging and asleep deep brain stimulator electrode placement, it is still a physiologic target and thus some form of intraoperative physiology is necessary. This chapter reviews the neurophysiologic monitoring method of microelectrode recording that is commonly employed during these neurosurgical procedures today.

Real-time removal of stimulation artifacts in closed-loop deep brain stimulation

Objective.Closed-loop deep brain stimulation (DBS) with neural feedback has shown great potential in improving the therapeutic effect and reducing side effects. However, the amplitude of stimulation artifacts is much larger than the local field potentials, which remains a bottleneck in developing a closed-loop stimulation strategy with varied parameters.Approach.We proposed an irregular sampling method for the real-time removal of stimulation artifacts. The artifact peaks were detected by applying a threshold to the raw recordings, and the samples within the contaminated period of the stimulation pulses were excluded and replaced with the interpolation of the samples prior to and after the stimulation artifact duration. This method was evaluated with both simulation signals andin vivoclosed-loop DBS applications in Parkinsonian animal models.Main results. The irregular sampling method was able to remove the stimulation artifacts effectively with the simulation signals. The relative errors between the power spectral density of the recovered and true signals within a wide frequency band (2-150 Hz) were 2.14%, 3.93%, 7.22%, 7.97% and 6.25% for stimulation at 20 Hz, 60 Hz, 130 Hz, 180 Hz, and stimulation with variable low and high frequencies, respectively. This stimulation artifact removal method was verified in real-time closed-loop DBS applicationsin vivo, and the artifacts were effectively removed during stimulation with frequency continuously changing from 130 Hz to 1 Hz and stimulation adaptive to beta oscillations.Significance.The proposed method provides an approach for real-time removal in closed-loop DBS applications, which is effective in stimulation with low frequency, high frequency, and variable frequency. This method can facilitate the development of more advanced closed-loop DBS strategies.

Improved detection of Parkinsonian resting tremor with feature engineering and Kalman filtering

OBJECTIVE

Accurate and reliable detection of tremor onset in Parkinson's disease (PD) is critical to the success of adaptive deep brain stimulation (aDBS) therapy. Here, we investigated the potential use of feature engineering and machine learning methods for more accurate detection of rest tremor in PD.

METHODS

We analyzed the local field potential (LFP) recordings from the subthalamic nucleus region in 12 patients with PD (16 recordings). To explore the optimal biomarkers and the best performing classifier, the performance of state-of-the-art machine learning (ML) algorithms and various features of the subthalamic LFPs were compared. We further used a Kalman filtering technique in feature domain to reduce the false positive rate.

RESULTS

The Hjorth complexity showed a higher correlation with tremor, compared to other features in our study. In addition, by optimal selection of a maximum of five features with a sequential feature selection method and using the gradient boosted decision trees as the classifier, the system could achieve an average F1 score of up to 88.7% and a detection lead of 0.52 s. The use of Kalman filtering in feature space significantly improved the specificity by 17.0% (p = 0.002), thereby potentially reducing the unnecessary power dissipation of the conventional DBS system.

CONCLUSION

The use of relevant features combined with Kalman filtering and machine learning improves the accuracy of tremor detection during rest.

SIGNIFICANCE

The proposed method offers a potential solution for efficient on-demand stimulation for PD tremor.

Parkinsonian Tremor Detection from Subthalamic Nucleus Local Field Potentials for Closed-Loop Deep Brain Stimulation

Deep Brain Stimulation (DBS) is a widely used therapy to ameliorate symptoms experienced by patients with Parkinson's Disease (PD). Conventional DBS is continuously ON even though PD symptoms fluctuate over time leading to undesirable side-effects and high energy requirements. This study investigates the use of a Iogistic regression-based classifier to identify periods when PD patients have rest tremor exploiting Local Field Potentials (LFPs) recorded with DBS electrodes implanted in the Subthalamic Nucleus in 7 PD patients (8 hemispheres). Analyzing 36.1 minutes of data with a 512 milliseconds non-overlapping window, the classification accuracy was well above chance-level for all patients, with Area Under the Curve (AUC) ranging from 0.67 to 0.93. The features with the most discriminative ability were, in descending order, power in the 31-45 Hz, 5-7 Hz, 21-30 Hz, 46-55 Hz, and 56-95 Hz frequency bands. These results suggest that using a machine learning-based classifier, such as the one proposed in this study, can form the basis for on-demand DBS therapy for PD tremor, with the potential to reduce side-effects and lower battery consumption.

Non-Linear Dynamical Analysis of Resting Tremor for Demand-Driven Deep Brain Stimulation

Parkinson's Disease (PD) is currently the second most common neurodegenerative disease. One of the most characteristic symptoms of PD is resting tremor. Local Field Potentials (LFPs) have been widely studied to investigate deviations from the typical patterns of healthy brain activity. However, the inherent dynamics of the Sub-Thalamic Nucleus (STN) LFPs and their spatiotemporal dynamics have not been well characterized. In this work, we study the non-linear dynamical behaviour of STN-LFPs of Parkinsonian patients using ε -recurrence networks. RNs are a non-linear analysis tool that encodes the geometric information of the underlying system, which can be characterised (for example, using graph theoretical measures) to extract information on the geometric properties of the attractor. Results show that the activity of the STN becomes more non-linear during the tremor episodes and that ε -recurrence network analysis is a suitable method to distinguish the transitions between movement conditions, anticipating the onset of the tremor, with the potential for application in a demand-driven deep brain stimulation system.

A machine-learning approach to volitional control of a closed-loop deep brain stimulation system

OBJECTIVE

Deep brain stimulation (DBS) is a well-established treatment for essential tremor, but may not be an optimal therapy, as it is always on, regardless of symptoms. A closed-loop (CL) DBS, which uses a biosignal to determine when stimulation should be given, may be better. Cortical activity is a promising biosignal for use in a closed-loop system because it contains features that are correlated with pathological and normal movements. However, neural signals are different across individuals, making it difficult to create a 'one size fits all' closed-loop system.

APPROACH

We used machine learning to create a patient-specific, CL DBS system. In this system, binary classifiers are used to extract patient-specific features from cortical signals and determine when volitional, tremor-evoking movement is occurring to alter stimulation voltage in real time.

MAIN RESULTS

This system is able to deliver stimulation up to 87%-100% of the time that subjects are moving. Additionally, we show that the therapeutic effect of the system is at least as good as that of current, continuous-stimulation paradigms.

SIGNIFICANCE

These findings demonstrate the promise of CL DBS therapy and highlight the importance of using subject-specific models in these systems.

Toward adaptive deep brain stimulation in Parkinson's disease: a review

Clinical deep brain stimulation (DBS) is now regarded as the therapeutic intervention of choice at the advanced stages of Parkinson's disease. However, some major challenges of DBS are stimulation induced side effects and limited pacemaker battery life. Side effects and shortening of pacemaker battery life are mainly as a result of continuous stimulation and poor stimulation focus. These drawbacks can be mitigated using adaptive DBS (aDBS) schemes. Side effects resulting from continuous stimulation can be reduced through adaptive control using closed-loop feedback, while those due to poor stimulation focus can be mitigated through spatial adaptation. Other advantages of aDBS include automatic, rather than manual, initial adjustment and programming, and long-term adjustments to maintain stimulation parameters with changes in patient's condition. Both result in improved efficacy. This review focuses on the major areas that are essential in driving technological advances for the various aDBS schemes. Their challenges, prospects and progress so far are analyzed. In addition, important advances and milestones in state-of-the-art aDBS schemes are highlighted – both for closed-loop adaption and spatial adaption. With perspectives and future potentials of DBS provided at the end.

Decoding Pigeon Behavior Outcomes Using Functional Connections among Local Field Potentials

Recent studies indicate that the local field potential (LFP) carries information about an animal's behavior, but issues regarding whether there are any relationships between the LFP functional networks and behavior tasks as well as whether it is possible to employ LFP network features to decode the behavioral outcome in a single trial remain unresolved. In this study, we developed a network-based method to decode the behavioral outcomes in pigeons by using the functional connectivity strength values among LFPs recorded from the nidopallium caudolaterale (NCL). In our method, the functional connectivity strengths were first computed based on the synchronization likelihood. Second, the strength values were unwrapped into row vectors and their dimensions were then reduced by principal component analysis. Finally, the behavioral outcomes in single trials were decoded using leave-one-out combined with the k-nearest neighbor method. The results showed that the LFP functional network based on the gamma-band was related to the goal-directed behavior of pigeons. Moreover, the accuracy of the network features (74 ± 8%) was significantly higher than that of the power features (61 ± 12%). The proposed method provides a powerful tool for decoding animal behavior outcomes using a neural functional network.

Parkinsonian rest tremor can be detected accurately based on neuronal oscillations recorded from the subthalamic nucleus

OBJECTIVE

To investigate the possibility of tremor detection based on deep brain activity.

METHODS

We re-analyzed recordings of local field potentials (LFPs) from the subthalamic nucleus in 10 PD patients (12 body sides) with spontaneously fluctuating rest tremor. Power in several frequency bands was estimated and used as input to Hidden Markov Models (HMMs) which classified short data segments as either tremor-free rest or rest tremor. HMMs were compared to direct threshold application to individual power features.

RESULTS

Applying a threshold directly to band-limited power was insufficient for tremor detection (mean area under the curve [AUC] of receiver operating characteristic: 0.64, STD: 0.19). Multi-feature HMMs, in contrast, allowed for accurate detection (mean AUC: 0.82, STD: 0.15), using four power features obtained from a single contact pair. Within-patient training yielded better accuracy than across-patient training (0.84vs. 0.78, p=0.03), yet tremor could often be detected accurately with either approach. High frequency oscillations (>200Hz) were the best performing individual feature.

CONCLUSIONS

LFP-based markers of tremor are robust enough to allow for accurate tremor detection in short data segments, provided that appropriate statistical models are used.

SIGNIFICANCE

LFP-based markers of tremor could be useful control signals for closed-loop deep brain stimulation.

Long-term detection of Parkinsonian tremor activity from subthalamic nucleus local field potentials

Current deep brain stimulation paradigms deliver continuous stimulation to deep brain structures to ameliorate the symptoms of Parkinson's disease. This continuous stimulation has undesirable side effects and decreases the lifespan of the unit's battery, necessitating earlier replacement. A closed-loop deep brain stimulator that uses brain signals to determine when to deliver stimulation based on the occurrence of symptoms could potentially address these drawbacks of current technology. Attempts to detect Parkinsonian tremor using brain signals recorded during the implantation procedure have been successful. However, the ability of these methods to accurately detect tremor over extended periods of time is unknown. Here we use local field potentials recorded during a deep brain stimulation clinical follow-up visit 1 month after initial programming to build a tremor detection algorithm and use this algorithm to detect tremor in subsequent visits up to 8 months later. Using this method, we detected the occurrence of tremor with accuracies between 68-93%. These results demonstrate the potential of tremor detection methods for efficacious closed-loop deep brain stimulation over extended periods of time.

A Fuzzy Inference System for Closed-Loop Deep Brain Stimulation in Parkinson's Disease

Parkinsons disease is a complex neurodegenerative disorder for which patients present many symptoms, tremor being the main one. In advanced stages of the disease, Deep Brain Stimulation is a generalized therapy which can significantly improve the motor symptoms. However despite its beneficial effects on treating the symptomatology, the technique can be improved. One of its main limitations is that the parameters are fixed, and the stimulation is provided uninterruptedly, not taking into account any fluctuation in the patients state. A closed-loop system which provides stimulation by demand would adjust the stimulation to the variations in the state of the patient, stimulating only when it is necessary. It would not only perform a more intelligent stimulation, capable of adapting to the changes in real time, but also extending the devices battery life, thereby avoiding surgical interventions. In this work we design a tool that learns to recognize the principal symptom of Parkinsons disease and particularly the tremor. The goal of the designed system is to detect the moments the patient is suffering from a tremor episode and consequently to decide whether stimulation is needed or not. For that, local field potentials were recorded in the subthalamic nucleus of ten Parkinsonian patients, who were diagnosed with tremor-dominant Parkinsons disease and who underwent surgery for the implantation of a neurostimulator. Electromyographic activity in the forearm was simultaneously recorded, and the relation between both signals was evaluated using two different synchronization measures. The results of evaluating the synchronization indexes on each moment represent the inputs to the designed system. Finally, a fuzzy inference system was applied with the goal of identifying tremor episodes. Results are favourable, reaching accuracies of higher 98.7% in 70% of the patients.