Brain Electrodes Research Articles

Learning from brain control: clinical application of brain–computer interfaces

Brain–computer interfaces (BCI) use neuroelectric and metabolic brain activity to activate peripheral devices and computers without mediation of the motor system. In order to activate the BCI patients have to learn a certain amount of brain control. Self-regulation of brain activity was found to follow the principles of skill learning and instrumental conditioning. This review focuses on the clinical application of brain–computer interfaces in paralyzed patients with locked-in syndrome and completely locked-in syndrome (CLIS). It was shown that electroencephalogram (EEG)-based brain–computer interfaces allow selection of letters and words in a computer menu with different types of EEG signals. However, in patients with CLIS without any muscular control, particularly of eye movements, classical EEG-based brain–computer interfaces were not successful. Even after implantation of electrodes in the human brain, CLIS patients were unable to communicate. We developed a theoretical model explaining this fundamental deficit in instrumental learning of brain control and voluntary communication: patients in complete paralysis extinguish goal-directed response-oriented thinking and intentions. Therefore, a reflexive classical conditioning procedure was developed and metabolic brain signals measured with near infrared spectroscopy were used in CLIS patients to answer simple questions with a “yes” or “no”-brain response. The data collected so far are promising and show that for the first time CLIS patients communicate with such a BCI system using metabolic brain signals and simple reflexive learning tasks. Finally, brain machine interfaces and rehabilitation in chronic stroke are described demonstrating in chronic stroke patients without any residual upper limb movement a surprising recovery of motor function on the motor level as well as on the brain level. After extensive combined BCI training with behaviorally oriented physiotherapy, significant improvement in motor function was shown in this previously intractable paralysis. In conclusion, clinical application of brain machine interfaces in well-defined and circumscribed neurological disorders have demonstrated surprisingly positive effects. The application of BCIs to psychiatric and clinical–psychological problems, however, at present did not result in substantial improvement of complex behavioral disorders.

Clinical study on deep brain electrode implantation assisted by ROSA system

Objective To evaluate the accuracy and safety of deep brain electrode implantation assisted by robotized stereotactic assistant (ROSA) system. Methods A total of 6 epileptic patients who had difficulty in positioning of epileptogenic focus underwent deep brain electrode implantation assisted by ROSA system. Record the maximum and minimum distance of electrode deviation according to postoperative CT and MRI and calculate the average distance, so as to evaluate the accuracy of operation. Record intraoperative electrode related bleeding events, postoperative electrode related bleeding or ischemic events and complications during the follow-up period, so as to evaluate the safety of operation. Results A total of 37 electrodes were successfully implanted in all 6 cases, with a success rate of 100%. Implanted electrodes were mainly located in medial temporal lobe, hippocampus and insular lobe (5 cases) and parieto-lobe (one case). The maximum deviation distance was 8.79 mm, and minimum was 1.14 mm, with an average of (3.65 ± 1.82) mm. Neither intraoperative electrode related bleeding event nor postoperative electrode related bleeding or ischemic event was found. No patient suffered from severe complications, such as intracranial infection, electrode disjunction, undesirable healing of scalp and wound infection, during the follow-up period from 3 to 5 months (mean 3.67 months). Conclusions Deep brain electrode implantation assisted by ROSA system is safe and accurate. DOI: 10.3969/j.issn.1672-6731.2015.09.006

Magnetic resonance-transcranial ultrasound fusion imaging: A novel tool for brain electrode location.

A combination of preoperative magnetic resonance imaging (MRI) with real-time transcranial ultrasound, known as fusion imaging, may improve postoperative control of deep brain stimulation (DBS) electrode location. Fusion imaging, however, employs a weak magnetic field for tracking the position of the ultrasound transducer and the patient's head. Here we assessed its feasibility, safety, and clinical relevance in patients with DBS. Eighteen imaging sessions were conducted in 15 patients (7 women; aged 52.4 ± 14.4 y) with DBS of subthalamic nucleus (n = 6), globus pallidus interna (n = 5), ventro-intermediate (n = 3), or anterior (n = 1) thalamic nucleus and clinically suspected lead displacement. Minimum distance between DBS generator and magnetic field transmitter was kept at 65 cm. The pre-implantation MRI dataset was loaded into the ultrasound system for the fusion imaging examination. The DBS lead position was rated using validated criteria. Generator DBS parameters and neurological state of patients were monitored. Magnetic resonance-ultrasound fusion imaging and volume navigation were feasible in all cases and provided with real-time imaging capabilities of DBS lead and its location within the superimposed magnetic resonance images. Of 35 assessed lead locations, 30 were rated optimal, three suboptimal, and two displaced. In two cases, electrodes were re-implanted after confirming their inappropriate location on computed tomography (CT) scan. No influence of fusion imaging on clinical state of patients, or on DBS implantable pulse generator function, was found. Magnetic resonance-ultrasound real-time fusion imaging of DBS electrodes is safe with distinct precautions and improves assessment of electrode location. It may lower the need for repeated CT or MRI scans in DBS patients.

Application of higher-order spectral analysis to local field potentials recorded in patients treated with deep brain stimulation.

Local field potentials (LFPs) recorded from implanted deep brain electrodes demonstrated the oscillatory nature of human basal ganglia. LFP rhythms were mainly characterized by means od power spectral analysis, thus loosing information related to rhythm phase synchronization and to event related phase modulations. Because the application of higher-order spectral analysis methodology can overcome such limitation, here we review the present applications of bispectral and cross-bispectral analysis to LFP recordings. The results obtained up to now showed that higher-order spectral analysis was able to clarify detect different rhythm synchronizations and interactions characterizing different pathologies and patient's states.

Localization of Metal Electrodes in the Intact Rat Brain Using Registration of 3D Microcomputed Tomography Images to a Magnetic Resonance Histology Atlas.

Simultaneous neural recordings taken from multiple areas of the rodent brain are garnering growing interest because of the insight they can provide about spatially distributed neural circuitry. The promise of such recordings has inspired great progress in methods for surgically implanting large numbers of metal electrodes into intact rodent brains. However, methods for localizing the precise location of these electrodes have remained severely lacking. Traditional histological techniques that require slicing and staining of physical brain tissue are cumbersome and become increasingly impractical as the number of implanted electrodes increases. Here we solve these problems by describing a method that registers 3D computed tomography (CT) images of intact rat brains implanted with metal electrode bundles to a magnetic resonance imaging histology (MRH) atlas. Our method allows accurate visualization of each electrode bundle’s trajectory and location without removing the electrodes from the brain or surgically implanting external markers. In addition, unlike physical brain slices, once the 3D images of the electrode bundles and the MRH atlas are registered, it is possible to verify electrode placements from many angles by “reslicing” the images along different planes of view. Furthermore, our method can be fully automated and easily scaled to applications with large numbers of specimens. Our digital imaging approach to efficiently localizing metal electrodes offers a substantial addition to currently available methods, which, in turn, may help accelerate the rate at which insights are gleaned from rodent network neuroscience.

Human implicit intent recognition based on the phase synchrony of EEG signals

This paper proposes a human implicit intent recognition system based on electroencephalography (EEG) signals, for developing an advanced interactive web service engine. We focus on identifying brain state transitions between intentions, and classifying a user's implicit intentions while viewing an image on a web page, based on an EEG experiment. We measure brain state changes between a navigational intention and an informational intention by using phase synchrony; i.e., the phase locking value (PLV) in an EEG. Comparing PLVs that correspond to the two intention states is useful for determining a human's implicit intention. In order to discriminate between a user's implicit intentions using a PLV, we must extract features based on an EEG analysis. For this purpose, we identify the most reactive band within the full band of brain signals. Theta (4–7 Hz), alpha (8–13 Hz), beta-1 (14–22 Hz), and beta-2 (23–30 Hz) bands are used to extract the EEG features from the most reactive EEG band. Subsequently, we select the most significant pairs (MSPs) that are highly reactive and correspond to the intention. According to the proposed method, these features are useful for: (i) showing the brain state transitions regarding intentions, and (ii) classifying a human's implicit intention using several classifiers such as a support vector machine (SVM), Gaussian Mixture Model (GMM), and Naïve Bayes. We then compare the results of these classifiers. This study demonstrates the potential uses of these identified brain electrode pairs for cognitive detection and task classification in future brain-computer interface (BCI) applications.

A spotlight on microneurotrophins: the future of amyotrophic lateral sclerosis treatment?

James P Bennett Jr* speaks to Daphne Boulicault, Commissioning Editor: James P Bennett, Jr is a native of St Petersburg, FL, USA. He received a BS in chemistry (with Honors) from the University of Florida (1970), then attended Johns Hopkins University School of Medicine for his MD (1974) and PhD (pharmacology, 1977) degrees. He worked in the laboratory of Dr Solomon Snyder, MD, for 6 years while in medical and graduate schools and trained in neuropharmocology. He then trained clinically for 2 years in internal medicine (1978–1980), followed by Neurology residency at the University of Virginia (UVA; 1980–1983) where he was Chief Resident (1982–1983). He became a faculty member in the UVA Departments of Neurology, Psychiatry and Pharmacology (1983–2009) and held the Ebbert Chair in Medical Science. In 2009, he moved to Virginia Commonwealth University, where he is the Bemiss Professor, was Chair of Neurology (2009–2013) and is founding Director of the Parkinson’s Disease and Movement Disorder Center (2009–2015). He has held multiple NIH grants and directed the NIH-funded Udall Parkinson’s Center at UVA. He investigates mitochondrial dysfunction in adult neurodegenerative diseases. He holds multiple patents related to experimental therapies of degenerative disorders.

Time Stability and Coherence Analysis of Multiunit, Single-Unit and Local Field Potential Neuronal Signals in Chronically Implanted Brain Electrodes

Introducing neural sensing and decoding to open-loop neurostimulation technologies has the potential to significantly improve the diagnosis and treatment of a wide variety of diseases treated through bioelectronic medicine. Chronically implanted multi-electrode arrays (MEA) can be used for such neural sensing and are critical for obtaining data of high spatial and temporal resolution to provide accurate decoding. Signals recorded from these arrays include local field potentials (LFP), and multiunit (MU) and single-unit (SU) activity. LFP offer signal stability over time, but at the expense of decreased spatial resolution. SU activity, on the other hand, offers better spatial resolution, but is considered less stable in chronic applications. MU activity, which represents an aggregate spiking activity of a population of neurons on the order of several hundred microns away from the recording tip, is considered a signal that can offer a compromise between the two signals. Here we used a wavelet decomposition method to extract and characterize the LFP, MU and SU signals obtained from a 96-channel MEA implanted in the motor cortex of a nonhuman primate over a 7.5-month period. We observed that not only are the MU signals more stable over time compared with SU activity, but that they are also significantly less correlated among electrodes compared with LFP over the spatial scale of the implanted array. Histological analysis of tissue sections also revealed a 51% reduction in the number of neuronal cell bodies within a radius around the electrode tips of the implanted tissue compared with control tissue. Our results indicate that MU activity offers long-term signal stability with less correlated signals, potentially providing an effective signal for neural sensing in bioelectronic medicine.

Subdural electrodes in focal epilepsy surgery at a typical academic epilepsy center.

To assess the use of subdural intracranial EEG (iEEG) on postoperative outcomes at an epilepsy center. Ninety-one epilepsy patients underwent iEEG. Array design was compared with noninvasive EEG with over 1 year of outcome. Patient demographics, implanted brain sites, electrodes, contacts/site, and surgical location were correlated with outcome. Fisher exact test and logistics regression were used to evaluate significance (P ≤ 0.05). Of ninety-one women, 55 (mean age, 32.3 years; range, 11-60) underwent tailored iEEG. Seventy of ninety-one (76.9%) resections (70% temporal) yielded 24/91 (26.4%) seizure free (SF). Strips (57.1%), grids (5.5%), or both (37.4%) for iEEG use was commonly bilateral (58.2%; 65.3% bitemporal) but did not predict outcome (P = NS). A lesion (28/91) did predict a SF outcome (42.9%). The iEEG localized 45.7% of seizures beyond scalp EEG and changed the localization or lateralization in 75.7% of resected patients. Electrode design, localization, lateralization, and site of resection did not correlate with outcome (P = NS). Overall, iEEG use portended a non-SF outcome (P ≤ 0.0001). The use of iEEG selected 46% additional patients for surgery, yet only 26% became SF. A magnetic resonance imaging lesion predicted a SF or seizure-improved outcome. Although iEEG changed the localization and lateralization of scalp ictal EEG in three quarters of patients, its use was a negative predictor for a favorable outcome. Preoperative counseling should emphasize expectations for seizure reduction in patients requiring iEEG.

Changes in cognitive abilities after deep brain stimulation for Parkinson disease.

Persons with Parkinson disease (PD) who have difficulties from side effects or dosing of medications, or both, have the option of undergoing deep brain stimulation (DBS). DBS provides electrical stimulation via electrodes implanted into one or both sides of the brain. These electrodes are connected to a programmable (adjustable) pulse generator (like a pacemaker) placed under the skin of the chest. The 2 most common targets for the brain electrodes are the subthalamic nuclei (STN) or globi pallidi (GPi). One of each of these targets is found in the left and right halves of the brain. These brain structures contain circuits important for motor functions or movement. Nearby in the brain are circuits that regulate emotion, mood, and cognition. Cognition refers to mental abilities such as memory, abstract thinking, and word finding. There is concern that the benefits of DBS for movement might be offset by negative effects of DBS on thinking and emotion. Studies have shown that DBS is safe from cognitive and emotional standpoints. Still, a minority of patients have problems with cognition and emotion after DBS. It is unclear whether stimulating the STN or GPi is safer. A few studies examining a small number of patients (and often using a limited number of tests) suggest that stimulation of the GPi might be safer than the STN for cognition and emotion.1 A large well-done study is needed to help physicians and patients make the best choices. This study by Odekerken and colleagues2 is 1 of 2 recently published that seek to address the important question of whether STN or GPi DBS is safer.

Stimulation frequency of deep brain electrodes in the subthalamic nucleus modifies response accurancy in a decision making task in Parkinson's disease

Stimulation frequency of deep brain electrodes in the subthalamic nucleus modifies response accurancy in a decision making task in Parkinson's disease

Correlation of the impedance and effective electrode area of doped PEDOT modified electrodes for brain-machine interfaces.

Electrode impedance is used to assess the thermal noise and signal-to-noise ratio for brain-machine interfaces. An intermediate frequency of 1 kHz is typically measured, although other frequencies may be better predictors of device performance. PEDOT-PSS, PEDOT-DBSA and PEDOT-pTs conducting polymer modified electrodes have reduced impedance at 1 kHz compared to bare metal electrodes, but have no correlation with the effective electrode area. Analytical solutions to impedance indicate that all low-intermediate frequencies can be used to compare the electrode area at a series RC circuit, typical of an ideal metal electrode in a conductive solution. More complex equivalent circuits can be used for the modified electrodes, with a simplified Randles circuit applied to PEDOT-PSS and PEDOT-pTs and a Randles circuit including a Warburg impedance element for PEDOT-DBSA at 0 V. The impedance and phase angle at low frequencies using both equivalent circuit models is dependent on the electrode area. Low frequencies may therefore provide better predictions of the thermal noise and signal-to-noise ratio at modified electrodes. The coefficient of variation of the PEDOT-pTs impedance at low frequencies was lower than the other conducting polymers, consistent with linear and steady-state electroactive area measurements. There are poor correlations between the impedance and the charge density as they are not ideal metal electrodes.

Thin-film electroencephalographic electrodes using multi-walled carbon nanotubes are effective for neurosurgery.

BackgroundIntraoperative morphological and functional monitoring is essential for safe neurosurgery. Functional monitoring is based on electroencephalography (EEG), which uses silver electrodes. However, these electrodes generate metal artifacts as silver blocks X-rays, creating white radial lines on computed tomography (CT) images during surgery. Thick electrodes interfere with surgical procedures. Thus, thinner and lighter electrodes are ideal for intraoperative use.MethodsThe authors developed thin brain electrodes using carbon nanotubes that were formed into thin sheets and connected to electrical wires.ResultsThe nanotube sheets were soft and fitted the curve of the head very well. When attached to the head using paste, the impedance of the newly developed electrodes was 5 kΩ or lower, which was similar to that of conventional metal electrodes. These electrodes can be used in combination with intraoperative CT, magnetic resonance imaging (MRI), or cerebral angiography. Somatosensory-evoked potentials, auditory brainstem responses, and visually evoked potentials were clearly identified in ten volunteers. The electrodes, without any artifacts that distort images, did not interfere with X-rays, CT, or MR images. They also did not cause skin damage.ConclusionsCarbon nanotube electrodes may be ideal for neurosurgery.

Electrophysiological registration of phonological perception in the subthalamic nucleus of patients with Parkinson’s Disease

Phonological processing is usually associated with the activation of cortical areas, especially in the left cerebral hemisphere. This study examined if phonologically elicited evoked potentials can be recorded directly from the subthalamic nucleus in patients with Parkinson’s Disease (PD). Seven PD patients who had undergone implantation of deep brain electrodes for the stimulation of the subthalamic nucleus were included. Local field potentials were recorded in a pre-attentive auditory phonological task, an attentive auditory phonological discrimination task, and a word recognition task. Auditory evoked potentials related to phonological, but not lexical processing, could be demonstrated in the subthalamic nucleus for all three tasks. Only minor changes were found after levodopa administration.This study demonstrates that the subthalamic nucleus is involved in early phonological perception, which puts the subthalamic nucleus in a position to modify phonological perception in a larger cortico-subcortical network.

Deep brain stimulation for movement disorders.

Deep brain stimulation (DBS) is an implanted electrical device that modulates specific targets in the brain resulting in symptomatic improvement in a particular neurologic disease, most commonly a movement disorder. It is preferred over previously used lesioning procedures due to its reversibility, adjustability, and ability to be used bilaterally with a good safety profile. Risks of DBS include intracranial bleeding, infection, malposition, and hardware issues, such migration, disconnection, or malfunction, but the risk of each of these complications is low--generally ≤ 5% at experienced, large-volume centers. It has been used widely in essential tremor, Parkinson's disease, and dystonia when medical treatment becomes ineffective, intolerable owing to side effects, or causes motor complications. Brain targets implanted include the thalamus (most commonly for essential tremor), subthalamic nucleus (most commonly for Parkinson's disease), and globus pallidus (Parkinson's disease and dystonia), although new targets are currently being explored. Future developments include brain electrodes that can steer current directionally and systems capable of "closed loop" stimulation, with systems that can record and interpret regional brain activity and modify stimulation parameters in a clinically meaningful way. New, image-guided implantation techniques may have advantages over traditional DBS surgery.

What limits the performance of current invasive brain machine interfaces?

The concept of a brain-machine interface (BMI) or a computer-brain interface is simple: BMI creates a communication pathway for a direct control by brain of an external device. In reality BMIs are very complex devices and only recently the increase in computing power of microprocessors enabled a boom in BMI research that continues almost unabated to this date, the high point being the insertion of electrode arrays into the brains of 5 human patients in a clinical trial run by Cyberkinetics with few other clinical tests still in progress. Meanwhile several EEG-based BMI devices (non-invasive BMIs) were launched commercially. Modern electronics and dry electrode technology made possible to drive the cost of some of these devices below few hundred dollars. However, the initial excitement of the direct control by brain waves of a computer or other equipment is dampened by large efforts required for learning, high error rates and slow response speed. All these problems are directly related to low information transfer rates typical for such EEG-based BMIs. In invasive BMIs employing multiple electrodes inserted into the brain one may expect much higher information transfer rates than in EEG-based BMIs because, in theory, each electrode provides an independent information channel. However, although invasive BMIs require more expensive equipment and have ethical problems related to the need to insert electrodes in the live brain, such financial and ethical costs are often not offset by a dramatic improvement in the information transfer rate. Thus the main topic of this review is why in invasive BMIs an apparently much larger information content obtained with multiple extracellular electrodes does not translate into much higher rates of information transfer? This paper explores possible answers to this question by concluding that more research on what movement parameters are encoded by neurons in motor cortex is needed before we can enjoy the next generation BMIs.

A step-wise approach to deep brain stimulation in mice

Studies of deep brain stimulation (DBS) in mice are rare due to their small size, agility, aversion to handling, and high anxiety compared to larger species. Studying DBS modulation of neural circuitry in murine models of human behavior may ensure safety, guide stimulatory parameters for clinical trials in humans, and inform a long-eluded mechanism. Stereotactic deep brain electrode implantation in a mouse is performed. Mechanical etching of the skull with a high-speed drill is used with placement of cyanoacrylate glue and molding of dental acrylate to affix the electrode in place. Stimulation experiments are conducted in the home cage after a habituation period. After testing is complete, electrode placement is verified in fixed tissue. Electrodes can be safely and accurately implanted in mice for DBS experimentation. Previous findings demonstrated accuracy in placement within the nucleus accumbens shell of 93 % [14]. In this study, there were no hardware malfunctions that required interrupting experimentation. Stereotactic DBS studies may be safely and effectively performed in mice to investigate neuropsychiatric disorders. In addition, examining the biochemical and molecular mechanisms underlying these disorders may be facilitated by widely available transgenic mouse lines and the Cre-Lox recombination system.

Mechanical Simulation of Neural Electrode-Brain Tissue Interface under Various Micromotion Conditions

Micromotion is one of the most important factors that influence the long-term stability of neural electrodes. In order to improve the long-term stability of brain-implanted electrodes, this study uses finite element simulation for static and transient analyses. In particular, the effects of micromotion frequency and adhesion state on the mechanical state of the electrode-brain tissue interface are investigated based on the commercial electrode A1x16-3-100-413. The results demonstrate that micromotion frequency has a great effect on the maximum von Mises (VM) stress, revealing that higher frequencies are more harmful than lower frequencies in terms of the long-term stability of the electrode. When the frequency is 20 Hz, the stress reaches its maximum value, and then increases very slightly with frequency increasing. Results also show that the degree of physical coupling between the electrode and the brain tissue has a significant influence on the interfacial mechanical state. Enhancing the adhesion between the electrode and the brain tissue can effectively decrease the stress, strain, and delamination of the microelectrode tip with respect to the adjacent neural tissue, and thus improve the working life of implanted electrodes. The cyclic hysteresis energy is highly dependent on the adhesion of the electrode-tissue interface; the energy values are highest for a low electrode-tissue physical coupling. The results indicate that it is important to minimize the micromotion of the brain tissue caused by neural electrodes. A neural electrode with a biofouling-resistant coating should thus be developed to optimize the interface to enhance the adhesion between the brain tissue and electrodes.

Nocturnal Paroxysmal Dystonia (NPD), two cases report

Nocturnal Paroxysmal Dystonia (NPD) is a rare condition characterized by recurrent attacks during NREM sleep of variable duration (seconds to minutes), with a complex clinical expression: repetitive stereotyped dystonic, ballistic or choreoathetoid movements involving single or all extremities and neck. It is associated with other conditions such as paroxysmal arousals, episodic nocturnal wanderings, dyskinetic or semipurposeful movements and vocalizations, often no ictal epileptic in surface EEG activity, and normal brain neuroimaging. Nowadays, it is clear that most of patients who fit the diagnosis of NPD have a form of sleep-activated focal epilepsy (nocturnal frontal lobe epilepsy). Our aim is to present two patients diagnosed with NPD. It is a descriptive study, enrolling two patients with NPD diagnosis. We report two cases. The first one, a 13 years old boy without relevant personal and family medical history, who debuted at 5 years old with dystonic and dyskinetic attacks during sleep that occurred many times per night. The second case is a 39 years old man, without relevant personal and family medical history. Since aged 31 years, he complained attacks of motor agitation with stereotypical movements every night and often several times per night. Both patients remained without neurological abnormalities in physical examination and brain neuroimaging. The EEGs during wakefulness and sleep were normal. Initially they were diagnosed as NREM parasomnia and followed many years without evidence of developing any neurocognitive disorder or neurological impairment, despite no treatment. The attacks of Nocturnal Paroxysmal Dystonia are a form of nocturnal frontal lobe epilepsy. The application of deep brain electrodes has led to the recognition that ”hypermotor seizures” may also be found in seizures originated from the temporal lobe and the insula. Functional brain imaging in frontal lobe seizures confirms that the peculiar motor patterns involve mesial and especially cingulate, motor areas. It is necessary to keep into consideration this rare and not well-known entity in the differential diagnosis of atypical parasomnias. Although the latest advances in Neurosciences, the polysomnography and the clinical history continue being pillars in the diagnosis of nocturnal frontal lobe epilepsy so the neurophysiological studies must always be considered when the diagnostic of a parasomnia is not well established. Dres. Quintana, Sáez y Paradinas.

Film-based Implants for Supporting Neuron-Electrode Integrated Interfaces for The Brain.

Neural engineering provides promise for cell therapy by integrating the host brain with brain-machine-interface technologies in order to externally modulate functions. Long-term interfaces with the host brain remain a critical challenge due to insufficient graft cell survivability and loss of brain electrode sensitivity over time. Here, integrated neuron-electrode interfaces were developed on thin flexible and transparent silk films as brain implants. Mechanical properties and surface topography of silk films were optimized to promote cell survival and alignment of primary rat cortical cells. Compartmentalized cultures of living neural circuit and co-patterned electrode arrays were incorporated on the silk films with built-in wire connections. Electrical stimulation via electrodes embedded in the films activated surrounding neurons evoked calcium responses. In mice brains the silk film implants showed conformal contact capable of modulating host brain cells with minimal inflammatory response and stable indwelling for weeks. The approach of combining cell therapy and brain electrodes could provide sustained functional brain-machine interfaces with ex vivo control of neuron-electrode interface with spatial and temporal precision.