A High Dynamic-Range Sensing Front-End For Neural Signal Recording Systems

Tolerates 650 mVpp CM and 80 mVpp DM artifacts Fabricated in 40 nm CMOS technologySignificantly improves Zin (5.3x) and the linear input range (2x)Increases the maximum resistance of duty-cycled resistors (32x)Does not require off-chip capsAchieves comparable power and noise performance

Closed-loop neuromodulationNeural stimulation therapy for drug-resistant neurological conditionsDiagnosis/treatment of neurological disorders, such as epileptic seizuresNeuroprosthetic technologies to aid in paralyzed patientsBrain-machine interfacesNeurological investigative tool

Researchers led by Professor Dejan Markovic have developed an innovative front-end technology that enables implantable neuromodulation systems to record and stimulate simultaneously. This neural recording chopper amplifier can tolerate 80 mVpp DM and 650 mVpp CM artifacts in a signal band of 1 Hz – 5 kHz, with a dynamic range of 74 dB in the spike band and 81 dB in the LFP band. This implantable system overcomes the limitations of wall-plugged systems, such as limited duration of recordings (hours vs. weeks), risk of infection (does not require an open wound), and patient discomfort. Moreover, this system will enable neuroscientists to probe and investigate the brain in ways that current technologies do not support.

State Of Development

Prototype chopper amplifiers have been fabricated in 40 nm CMOS technology with an area of 0.069 mm²/ch, using a total power of 2.8 µW from a 1.2 V core supply.

Background

Closed-loop neuromodulation with simultaneous stimulation and sensing is needed to administer therapy in patients suffering from drug-resistant neurological ailments. However, stimulation generates large artifacts at the recording sites, which saturate traditional front-ends with common-mode (CM) artifacts around 500 mV and differential-mode (DM) artifacts between 50 mV – 100 mV. Although power and noise limitations have been improved in prior work, these methods still suffer from low Z_in and limited input signal range, making them incapable of performing true closed-loop operation.

Additional Technologies by these Inventors

• Saturation-Tolerant Electrophysiological Recording Interface
• Scalable Parameterized VLSI Architecture for Compressive Sensing Sparse Approximation
• A Simple, Area-Efficient Ripple-Rejection Technique for Chopped Bio-Signal Amplifiers
• Autonomous Thermoelectric Energy-Harvesting Platform for Biomedical Sensors
• Load Adaptive, Reconfigurable Active Rectifier for Multiple Input Multple Output (MIMO) Implant Power Management
• Electrode Agnostic, Supply Variant Stimulation Engine For Implantable Neural Stimulation
• A Distance-Immune Low-Power Inductively-Coupled Bidirectional Data Link

Tech ID/UC Case

28914/2017-229-0

Related Cases

2017-229-0

美国

桌面版