A review on microelectrode recording selection of features for machine learning in deep brain stimulation surgery for Parkinson’s disease

ObjectiveThis study seeks to systematically review the selection of features and algorithms for machine learning and automation in deep brain stimulation surgery (DBS) for Parkinson’s disease. This will assist in consolidating current knowledge and accuracy levels to allow greater understanding and research to be performed in automating this process, which could lead to improved clinical outcomes.MethodsA systematic literature review search was conducted for all studies that utilized machine learning and DBS in Parkinson’s disease.ResultsTen studies were identified from 2006 utilizing machine learning in DBS surgery for Parkinson’s disease. Different combinations of both spike independent and spike dependent features have been utilized with different machine learning algorithms to attempt to delineate the subthalamic nucleus (STN) and its surrounding structures.ConclusionThe state-of-the-art algorithms achieve good accuracy and error rates with relatively short computing time, however, the currently achievable accuracy is not sufficiently robust enough for clinical practice. Moreover, further research is required for identifying subterritories of the STN.SignificanceThis is a comprehensive summary of current machine learning algorithms that discriminate the STN and its adjacent structures for DBS surgery in Parkinson’s disease.     

Systemic effects of deep brain stimulation on synergic control in Parkinson’s disease

Objective

We explored effects of deep brain stimulation (DBS) in patients with Parkinson’s disease (PD) on the synergic control of fingers in a multi-finger force production task and of muscles in a task involving vertical posture.

Combined deep brain stimulation and thalamotomy for tremor-dominant Parkinson’s disease

Although deep brain stimulation (DBS) is an established treatment for Parkinson’s disease, the long-term suppression of tremor is still a challenging issue. We report two patients with tremor-dominant Parkinson’s disease (PD) treated with unilateral thalamotomy of the ventralis intermedius nucleus (Vim) combined with the subthalamic nucleus (STN)-DBS or the posterior subthalamic area (PSA)-DBS. One year after the surgery, thalamotomy of the area from the Vim to the PSA showed improvement not only in tremor but also in rigidity and akinesia. PSA- or STN-DBS with low intensity stimulation eliminated residual PD symptoms. Combined DBS and thalamotomy may provide long-term improvement of the majority of PD symptoms using lower therapeutic stimulation voltages.     

Effect of subthalamic deep brain stimulation on non-motor fluctuations in Parkinson’s disease

The non-motor consequences of subthalamic stimulation are largely questioned. Cognition, motivation, anxiety, depression and even occurrence of suicides have been considered as a potential consequence of the surgical intervention. Non-motor fluctuations are present in all the patients with motor fluctuations and may sometimes be even more invalidating. Interestingly, subthalamic deep brain stimulation alleviates non-motor fluctuations allowing strikingly successful effects on sensory, dysautonomic and cognitive fluctuations while psychic fluctuations respond less consistently to this treatment. Nevertheless, severe mood fluctuations, oscillating from Off dysphoria to ON hypomania, are frequently associated with addictive behaviors and improve dramatically after subthalamic stimulation. This may be a further argument to support the indication of surgery for these patients.     

Self-adjustment of deep brain stimulation delays optimization in Parkinson’s disease

BackgroundParkinson’s Disease patients undergo time-consuming programming to refine stimulation parameters after deep brain stimulation surgery.ObjectiveTo assess whether the use of the advanced functions of a patient’s programmer would facilitate programming of deep brain stimulation.MethodsThirty patients were randomly allocated to the use of advanced versus simple mode of the patient programmer in this single-centre, prospective, randomized, controlled study. Primary outcome was the number of days required to optimize the stimulation settings.ResultsThe number of days required to optimize stimulation was significantly lower in the simple mode (88.5 ± 33.1 vs. 142.1 ± 67.4, p = 0.01). In addition, the advanced mode group had a higher number of side effects (5.4 ± 3.1 vs. 2.6 ± 1.9, p = 0.0055).ConclusionsThe use of the advanced functions of patient programmer delays programming optimization and it is associated with a higher number of side effects. These findings highlight the need for other methods for faster and safer stimulation programming.     

Subthalamic deep brain stimulation affects heading perception in Parkinson’s disease

Journal of Neurology - Parkinson’s disease (PD) presents with visuospatial impairment and falls. It is critical to understand how subthalamic deep brain stimulation (STN DBS) modulates...     

Deep brain stimulation in Parkinson’s disease

Throughout the past decade, there has been a marked increase in surgical therapies, primarily deep brain stimulation (DBS), for the treatment of advanced Parkinson’s disease (PD). DBS of the thalamus has been shown to be effective in reducing parkinsonian tremor; however, it is not the treatment of choice for PD given the progression of other symptoms such as rigidity and bradykinesia. Stimulation of the globus pallidus or the subthalamic nucleus is safe and efficacious in the long-term treatment of all cardinal symptoms of PD, and they are currently the surgeries of choice. Serious adverse events with DBS can occur in 1% to 2% of patients, infection in 5% to 8% of patients, and hardware complications in approximately 25% of patients. Complications associated with DBS are related to the experience of the surgical center. Referring physicians and patients should be aware of the number of surgical procedures and complication rates of any prospective surgical center.     

The impact of deep brain stimulation on the nonmotor symptoms of Parkinson’s disease

Deep brain stimulation (DBS) is a well-established therapy for patients with advanced Parkinson’s disease (PD) with clear benefits on many of the motor symptoms. The effects of DBS on the nonmotor symptoms are less well examined. Emergence of tools to measure the nonmotor burden in PD is now allowing a more objective assessment of impact of DBS on such symptoms. Here we review the pertinent evidence and conclude that, as a therapy, DBS has a major potential to contribute towards the holistic care of PD patients.     

Behavioral changes associated with deep brain stimulation surgery for Parkinson’s disease

Evaluation for psychiatric and cognitive dysfunction is a crucial part of preoperative assessment for deep brain stimulation (DBS) surgery. All candidates for DBS should be treated for active psychiatric disorders, and the treatment team should also screen for past psychiatric history, which may predispose to postoperative psychiatric illness. A wide range of behavioral symptoms have been seen following DBS, including depression, hallucinations or true psychosis, mania, and impulsivity. Suicidal ideation should be treated aggressively. Cognitive changes may occur, but data are currently too limited to make predictions regarding subtle effects on cognition. However, patients with dementia may experience further decline in cognitive function following DBS.     

Low-frequency subthalamic nucleus deep brain stimulation for axial symptoms in advanced Parkinson’s disease

Axial symptoms such as freezing of gait and falls are common manifestations of advanced Parkinson’s disease (PD) and are partially responsive to medical treatment. High-frequency (≥130 Hz) deep brain stimulation (DBS) of the subthalamic nucleus (STN) is highly efficacious in ameliorating appendicular symptoms in PD. However, it is typically less effective in improving axial symptomatology, especially in the long term. We have studied the effects of low-frequency stimulation (LFS) (≤80 Hz) for improving speech, gait and balance dysfunction in the largest patient population to date. PD patients with bilateral STN-DBS and resistant axial symptoms were switched from chronic 130 Hz stimulation to LFS and followed up to 4 years. Primary outcome measures were total motor UPDRS scores, and axial and gait subscores before and after LFS. Bivariate analyses and correlation coefficients were calculated for the different conditions. Potential predictors of therapeutic response were also investigated. Forty-five advanced PD patients who had high frequency stimulation (HFS) for 39.5 ± 27.8 consecutive months were switched to LFS. LFS was kept on for a median period of 111.5 days before the assessment. There was no significant improvement in any of the primary outcomes between HFS and LFS, although a minority of patients preferred to be maintained on LFS for longer periods of time. No predictive factors of response could be identified. There was overall no improvement from LFS in axial symptoms. This could be partly due to some study limitations. Larger prospective trials are warranted to better clarify the impact of stimulation frequency on axial signs.     

The potential neuromodulatory impact of subthalamic nucleus deep brain stimulation on Parkinson’s disease progression

IntroductionSTN-DBS has been claimed to change progression symptoms in animal models of PD, but information is lacking about the possible neuromodulatory role of STN-DBS in humans. The aim of this prospective controlled study was to evaluate the long-term impact of STN-DBS on motor disabilities and cognitive impairment in PD patients in comparison to Best-Medical-Therapy (BMT) and Long-term-Post-Operative (POP) groups.Material and methodsPatients were divided into 3 groups: the BMT-group consisted of 20 patients treated only with pharmacotherapy, the DBS-group consisted of 20 PD patients who underwent bilateral STN-DBS (examined pre- and postoperatively) and the POP-group consisted of 14 long-term postoperative patients in median 30 month-time after DBS. UPDRS III scale was measured during 3 visits in 9 ± 2 months periods (V1, V2, V3) in total-OFF phase. Cognitive assessment was performed during each visit in total-ON phase.ResultsThe comparable UPDRS III OFF gain was observed in both BMT-group and POP-group evaluations (p < 0.05). UPDRS III OFF results in DBS-group revealed significant UPDRS III OFF increase in ΔV2-V1 assessment (p < 0.05) with no significant UPDRS III OFF alteration in ΔV3-V2 DBS-group evaluation (p > 0.05). Cognitive assessment revealed significant alterations between DBS-group and BMT-group in working memory, executive functions and learning abilities (p < 0.05).ConclusionsThe impact of STN-DBS on UPDRS III OFF score and cognitive alterations suggest its neuromodulatory role, mainly during the first 9–18 months after surgery.     

Postoperative delirium in Parkinson’s disease patients following deep brain stimulation surgery

Deep brain stimulation (DBS) surgery is an effective treatment for patients with advanced Parkinson’s disease. Delirium in hospitalized Parkinson’s disease patients is common and often leads to prolonged hospital stays. This study reports on the incidence and etiology of postoperative delirium following DBS surgery. Patients (n = 59) with advanced Parkinson’s disease underwent bilateral (n = 56) or unilateral (n = 3) DBS electrode implant surgery, followed 1 week later with surgical placement of DBS generators. The development of delirium during either hospital stay was evaluated retrospectively from the hospital chart. Potential causes of delirium were evaluated, including history of delirium, opiate equivalents, medication administration delays and missed doses during hospitalization, and Parkinson’s disease duration. Delirium following implantation of DBS electrodes was common (22% of patients). It was less commonly associated with generator placement (10%). A history of delirium, age, and disease duration were positive predictors of delirium. Opiate equivalent doses were negatively correlated with delirium. Missed Parkinson’s medication doses (53% of patients) and delayed administration (81% of patients) were common, and had a slight relation with delirium. Delirium was not related to complexity of medication regimen or use of dementia medications. Despite the presence of delirium most patients still only required a single night in the hospital post-surgery (67%). Prolonged hospital stay was due not only to delirium but also severe off states and other medical issues. Recognition and expectant management of delirium is best accomplished in a multidisciplinary setting, including the patient’s family and nursing, pharmacy and neurological surgery staff.     

The treatment of Parkinson’s disease with deep brain stimulation：current issues

Deep brain stimulation has become a well-established symptomatic treatment for Parkinson’s disease during the last 25 years. Besides improving motor symptoms and long-term motor com-plications, positiv...     

Understanding the impact of deep brain stimulation on ambulatory activity in advanced Parkinson’s disease

Whilst deep brain stimulation of the subthalamic nucleus (DBS-STN) improves the motor symptoms of Parkinson's disease (PD), its effect on daily activity is unknown. We aimed to quantify changes in ambulatory activity following DBS-STN in advanced PD using novel accelerometry based measures that describe changes to the volume and pattern of walking. Seventeen participants with advanced PD were measured over a 7-day period using an activPAL (?) activity monitor. Data were collected 6 weeks before and 6 months after surgery and included measures that describe the volume and pattern of ambulatory activity (number of steps per day, accumulation, diversity and variability of walking time), alongside standard measures for disease severity, freezing of gait, gait speed, and extended activities of daily living. Activity outcomes were compared pre- and 6 months post-surgery using linear mixed models and correlated with standard outcomes. The results of this study are despite significant improvements in motor symptoms after surgery, the volume of ambulatory activity (total number of steps per day) did not change (P = 0.468). However, significant increases in length and variability of walking bouts emerged, suggesting improvements in diversity and flexibility of walking patterns. Motor severity and extended activities of daily living scores were significantly correlated with walking bout variability but not with volume of walking. Thus, the conclusions are reduction in motor symptom severity after DBS-STN translated into selective improvements in daily activity. Novel measures derived from accelerometry provide a discrete measure of performance and allow closer interpretation of the impact of DBS-STN on real-world activity.     

Differentiated effects of deep brain stimulation and medication on somatosensory processing in Parkinson’s disease

ObjectivesDeep brain stimulation (DBS) and dopaminergic medication effectively alleviate the motor symptoms in Parkinson’s disease (PD) patients, but their effects on the sensory symptoms of PD are still not well understood. To explore early somatosensory processing in PD, we recorded magnetoencephalography (MEG) from thirteen DBS-treated PD patients and ten healthy controls during median nerve stimulation.MethodsPD patients were measured during DBS-treated, untreated and dopaminergic-medicated states. We focused on early cortical somatosensory processing as indexed by N20m, induced gamma augmentation (31–45 Hz and 55–100 Hz) and induced beta suppression (13–30 Hz). PD patients’ motor symptoms were assessed by UPDRS-III.ResultsUsing Bayesian statistics, we found positive evidence for differentiated effects of treatments on the induced gamma augmentation (31–45 Hz) with highest gamma in the dopaminergic-medicated state and lowest in the DBS-treated and untreated states. In contrast, UPDRS-III scores showed beneficial effects of both DBS and dopaminergic medication on the patients’ motor symptoms. Furthermore, treatments did not affect the amplitude of N20m.ConclusionsOur results suggest differentiated effects of DBS and dopaminergic medication on cortical somatosensory processing in PD patients despite consistent ameliorating effects of both treatments on PD motor symptoms.SignificanceThe differentiated effect suggests differences in the effect mechanisms of the two treatments.     

A decision tool to support appropriate referral for deep brain stimulation in Parkinson’s disease

Background and objective

Although Deep Brain Stimulation (DBS) has been proven to be an effective treatment for patients with advanced Parkinson’s disease (PD), it may be difficult for general neurologists to identify appropriate candidates for this procedure. We developed an electronic decision tool that can assist neurologists in deciding which PD patients should be referred for DBS consideration.

Methods

Using the RAND/UCLA Appropriateness Method, an international expert panel assessed the appropriateness of referral for 972 theoretical patient profiles. Panel results were embedded in an electronic decision support tool which displays the panel statement on referral (appropriate, inappropriate and uncertain) after completion of the patient profile.

Results

Referral was considered appropriate for 33 % of the theoretical profiles. Logistic regression showed excellent internal consistency of the ratings (predictive value 92 %). Symptom severity (OFF-symptoms, dyskinesias, refractory tremor) and PD duration were positively associated with the panel judgment that referral is appropriate. Presence of levodopa-resistant axial symptoms, age ≥ 70 years and presence of cognitive impairment showed the strongest negative impact.

Conclusions

The RAND/UCLA method proved to be useful in determining the appropriate criteria for DBS referral. Validity and applicability of the decision tool (accessible via http://test.stimulus-dbs.org) in clinical practice need to be further determined.

     

A tool to improve pre-selection for deep brain stimulation in patients with Parkinson’s disease

Determining the eligibility of patients with Parkinson’s disease (PD) for deep brain stimulation (DBS) can be challenging for general (non-specialised) neurologists. We evaluated the use of an online screening tool (Stimulus) that aims to support appropriate referral to a specialised centre for the further evaluation of DBS. Implementation of the tool took place via an ongoing European multicentre educational programme, currently completed in 15 DBS centres with 208 referring neurologists. Use of the tool in daily practice was monitored via an online data capture programme. Selection decisions of patients referred with the assistance of the Stimulus tool were compared to those of patients outside the screening programme. Three years after the start of the programme, 3,128 patient profiles had been entered. The intention for referral was made for 802 patients and referral intentions were largely in accordance with the tool recommendations. Follow-up at 6 months showed that actual referral took place in only 28%, predominantly due to patients’ reluctance to undergo brain surgery. In patients screened with the tool and referred to a DBS centre, the acceptance rate was 77%, significantly higher than that of the unscreened population (48%). The tool showed a sensitivity of 99% and a specificity of 12% with a positive and negative predictive value of 79 and 75%, respectively. The Stimulus tool is useful in assisting general neurologists to identify appropriate candidates for DBS consideration. The principal reason for not referring potentially eligible patients is their reluctance to undergo brain surgery.