Electrical neuromodulation is an evolving therapy with the targeted delivery of constant voltage stimulation (CVS) or constant current stimulation (CCS) to specific neurological sites in a body to enable the alteration of nerve activities. It can influence nerves by releasing transmitters such as dopamine, or other chemical messengers such as the peptide substance, that can modulate the excitatory functions or inhibitory functions of neural circuits. The end effect is the normalization of a neural network function from its perturbed state. Presumed mechanisms of electrical neuromodulation could include a depolarizing blockade, stochastic normalization of neural firing, an axonal blockade, reduction of neural firing keratosis, and suppression of neural network oscillations. Although the exact mechanisms of electrical neuromodulation are not known, its empirical effectiveness has led to considerable clinical applications.
Electrical neuromodulation using implantable devices was first achieved in the 1980s. Its technologies and applications have continued to develop and expand. Existing and emerging electrical neuromodulation treatments have been applied to neural disorders like drug-resistant epilepsy, chronic pain, and Parkinson’s disease. They have also been used to improve sensory deficits, such as cochlear implants for hearing and retinal implants for vision, as well as functional therapies such as bladder or bowel control. Electrical neuromodulation therapy has been investigated for other chronic conditions, such as dementia, Alzheimer’s disease, depression, and rheumatoid arthritis. It is also called bioelectronic medicine, which can “turn off” chronic diseases or disorders by electricity. Bioelectronic medicine has great potential to be widely used as the future medicine.
In general, implantable electrical neuromodulation systems consist of electrodes and an implanted pulse generator (IPG) with associated external components. Electrodes can be epidural, subdural, or parenchymal electrodes placed via minimally invasive needle techniques, open-surgical exposure to the target, or stereotactic implants for the central nervous system. Depending on the distance from the electrode access point, an extension cable may be added to the system. The IPG can have either a non-rechargeable battery needing replacement every 2 – 5 years depending on the stimulation parameters, or a rechargeable battery that is replenished via an external inductive charging system.
Although most systems are operated in an open loop through the delivery of a constant train of stimulation, we are now at the advent of so-called closed-loop or feedforward stimulation, where the device’s activation is contingent on a physiological event, such as an epileptic seizure. In this circumstance, the device is activated to deliver a desynchronizing pulse to the cortical area that is undergoing an epileptic seizure. This concept of closed-loop stimulation is likely to become more prevalent as physiological markers of targeted diseases and neural disorders are discovered and verified [2]. On-demand closed-loop stimulation may contribute to a longer battery life if the sensing and signal-processing demands of the system are sufficiently power-efficient. New electrode designs could yield more efficient and precise stimulation, requiring less current and minimizing unwanted side-stimulation. The wireless power transfer to recharge the implanted battery and wireless bidirectional data transceivers to communicate with implanted devices have been adopted in neuromodulation systems.
In addition, to leverage the advanced CMOS nano-electronics technologies, the operation speed of algorithm with complex functions in the implanted medical devices to judge the symptoms of neuro-disorders can be processed quickly. Thus, the in-time treatment with the electrical stimulation can be executed to stop the symptoms of neuro-disorders. As well as, the power consumption of the SoC fabricated by the nanoscale CMOS technologies can be significantly reduced to get a longer battery life for the implanted medical device in medical field applications. Thus nano-bioelectronic medicine will become more and more attractive.
Biomedical Electronics Translational Research Center (BETRC)
Biomedical Electronics Translational Research Center (BETRC) at NYCU established in 2004 focuses on researches and developments of implantable medical electronic systems using System-on-a-Chip (SoC) technology and biocompatible materials, especially for close-loop neuromodulation for the treatment of neurological diseases through electrical voltage/current stimulations. The mission and vision of this BETRC include (1) to treat the intractable neurological disorders by conducting inter-disciplinary researches to develop multi-disciplinary technology platform, (2) to explore the frontier of neural sciences, and (3) to incubate the start-up companies on neural prosthetic devices. The technologies for biomedical devices and diagnosis equipment require many semiconductor chips and biomaterials can be developed or manufactured in Taiwan with strong industrial support. Fig. 1 illustrates the relationship of medical applications and technology platforms in the BETRC. The group photo of research team in BETRC is shown in Fig. 2.
There were five project teams in BETRC working on the following topics: (1) Photovoltaic – powered subretinal prostheses for patients with retinitis pigmentosa (RP) and age-related macular degeneration (AMD), (2) Closed-loop epileptic seizure control systems for patients with epilepsy or dementia , (3) Bone-guided cochlear prostheses for patients with high-frequency hearing loss, (4) Closed-loop Parkinson deep brain stimulation (DBS) system for Parkinson’s disease (PD), and (5) EEG(electroencephalogram)-tDCS(transcranial direct-current stimulation) system for neurological rehabilitation of stroke patients. Moreover, the BETRC invites partner universities to form the integrated engineering-biomedical joint research platform in Taiwan. The BETRC center also promotes international collaborations with worldwide top research centers, institutions, and enterprises.
The research projects have been partially supported by “the Aim for Top University Project” of the Ministry of Education (MoE), Taiwan from 2006-2015 and “the Featured Areas Research Center Program within the framework of the Higher Education Sprout Project” of MoE since 2015. Moreover, partial long-term supports are also from the National Science Council (NSC) and later the Ministry of Science and Technology (MOST), as well as the National Chiao Tung University (Now the National Yang-Ming Chiao-Tung University).
In 2013, the research work on the closed-loop epileptic seizure control SoC was published at the International Solid-State Circuits Conference (ISSCC) and awarded by the ISSCC 2013 Distinguished Technical Paper Award. In 2016, a start-up company called A-Neuron Electronic Corp. was established by transferring the technologies of BETRC on Photovoltaic-powered subretinal prostheses and closed-loop epileptic seizure control systems. The teams at A-Neuron Electronic Corp. will commercialize the technologies into two major medical products.
Fig. 3 shows the chip photographs of both closed-loop epileptic seizure control SoC and its external control system fabricated by 0.18-µm CMOS process technology [3]. The highlighted key achievements of SoC includes a low-noise ECoG (Electrocorticography) acquisition unit with input protection circuit to share electrodes with stimulators, a bio-signal processor for accurate and fast human seizure detection, a biphasic CCS up to 3mA with adaptive power supply, and a wireless power and bi-directional data telemetry in single pair of coils. When the epileptic seizure onset is detected, the biphasic CCS is enabled to stimulate the onset site of brain to stop the onset seizure before it spreads out to cause a large seizure. The wireless power transfer is used for battery charging and the bi-directional data telemetry is used to transfer ECoG out and control signal in. This is the first SoC for human closed-loop epileptic seizure control.
Intelligent Adaptive Closed-Loop DBS System for Parkinson’s Disease
Parkinson’s disease (PD) is a neurodegenerative disease of the motor nervous system. It is estimated that the prevalence of the PD is about 1% in the population over 60 years old. The pathological evidences showed that degeneration of dopaminergic neurons in the substantia nigra is the essential feature of the PD. Substantia nigra is a part of the basal ganglia, which plays a crucial role in motor control of voluntary movement. When the dopaminergic neurons of the substantia nigra degenerated, the balance between the direct and indirect pathways of the basal ganglia is disrupted, leading to symptoms of the PD. Patients may experience symptoms such as resting tremor, rigidity, and bradykinesia [5]. Medical treatment can alleviate related symptoms; however, long-term medication is susceptible to levodopa-induced dyskinesia, motor fluctuation, or other side effects.
For patients with advanced Parkinson’s disease, they may choose deep brain stimulation (DBS) to treat the drug-induced adverse effects. The conventional DBS (cDBS) is an open-loop system. When the power of medical device is turned on, the stimulator provides continuous electrical stimulation in the brain until the power is exhausted; usually about ~5 years later, the patient must undergo another surgery to replace the stimulator. Continuous electrical stimulation may cause some adverse effects, such as gait instability, dysarthria, or stimulation-induced dyskinesia. Because continuous electrical stimulation not only inhibits abnormal neural function, but also interferes normal neural function. Can it be possible to detect unique disease biomarker as feedback control signal for electrical stimulation? This is the concept of closed-loop system, in which physicians and scientists are highly interested.
The local field potential (LFP) is the ensemble activity of the synaptic potentials surrounded the lead contacts and can reflect the synchronized activity of a population of neurons. LFP can be recorded in the subthalamic nucleus (STN) of the PD patients when DBS leads were implanted in the operating room [6]. After the spectral analysis, abnormal beta-band oscillations (13-35 Hz) were observed in the STN of the patients [7]. This finding was similar to the recording in the internal globus pallidus (GPi) of the parkinsonian rhesus monkey [8]. When patients received medical treatment or cDBS on the STN, the beta oscillations were significantly inhibited and the symptoms of PD were also improved. Therefore, abnormal LFP of the STN can be used as a biomarker for the symptoms of PD, which has been widely recognized by experts.
If this unique beta oscillation is used as a feedback control signal for electrical stimulation, a proposed threshold can be set according to the suitable signal power. Professor Brown’s group first demonstrated that using beta oscillation as a feedback control signal for electrical stimulation within one week after DBS surgery in PD patients, their motor symptoms improvement score was significantly better than that of continuous electrical stimulation (50 % vs. 29%, p = 0.005), and the overall required electrical stimulation time was only 44% of the continuous electrical stimulation [9]. Custom-built device was used to design different algorithms of beta oscillation for closed-loop stimulation. According to the dynamic power change of beta oscillation, the stimulation intensity of the stimulator was adjusted. The motor symptoms were significantly improved compared with those in fixed-intensity electrical stimulation [10]. Such a closed-loop DBS, also known as the adaptive DBS (aDBS), has passed clinical trials for proof of principle.
The conventional devices used in adaptive DBS technology were implemented by system-on-board design. The latest aDBS device (Medtronic Summit RC+S) also adopts system-on-board design with several integrated circuit components of differential functions via connections through PCB board. In our work, the SoC-based design can improve the performance of detection, computing, and stimulation functions. Moreover, it can effectively reduce the operating power consumption of the implanted medical device. The microphotography of SoC developed with adaptive deep-brain detection and stimulation for implantable medical devices is shown in Fig. 4. Through integration co-design with the neural-signal acquisition analog frond-end circuit and the high-voltage tolerance stimulator circuit, the artifact interference during electrical stimulation can be successfully blocked from the acquired neural signals. In this SoC, it can simultaneously record the LFP signals from 16 channels within the implanted electrodes, converted them to digital signals, and then processed by bio-signal processor. The algorithm to evaluate the symptoms of PD, that co-developed with medical doctors, can be real-time processed by the bio-signal processor to judge whether the electrical stimulation will be applied, or not. Therefore, the closed-loop/adaptive deep brain stimulation can be efficiently executed by this developed SoC. The adaptive DBS implemented with SoC chip can increase the service life and therapeutic effect of the implanted stimulators.
A biomedical device built with the fabricated SoC chip has been prepared for pre-clinical tests, including animal experiments, electrical safety, recognized medical-device standards tests, and algorithm/software validation. After the safety certification and in vivo safety tests have been completed, clinical trials on human will be executed.
Prospective
Both implantable closed-loop epileptic seizure control systems and intelligent adaptive closed-loop DBS System for Parkinson’s disease in this article are physics-based circuits and systems. The essential design is on CMOS SoCs or integrated circuits. In the future development of bioelectronic medicines, it is believed that more and more advanced CMOS nano-electronics technologies, innovative circuit/system designs, and nano-bioelectronics technologies will be applied. These advancements will greatly benefit patients and lead to progress in neural science and engineering.