Accurate movement trajectory decoding for high-performance brain-machine interfaces
Brain-machine interfaces (BMIs) hold the promise to enable a direct communication channel between a paralyzed patient’s brain and external environment. Implementing modern machine learning algorithms for continuous movement intention decoding in a BMI is an interesting problem. While the classification of various motor tasks, such as binary classification of hand opening/closing or detecting moving fingers has been demonstrated in several studies, accurate decoding of continuous movement trajectory (i.e., position of hands/fingers) remains a challenge.
In this project, we aim to build a regressor to decode movement trajectory from brain recordings in patients (ECoG and/or spiking activity). Students will study informative biomarkers and implement ML models for target regression tasks to obtain a high decoding performance (both feature-engineered and end-to-end models).
Objectives: 1. Understanding of the basic problems and research directions in neural engineering. 2. Applying machine learning algorithms to a real world (and challenging) application. 3. Gaining experience in biomedical signal processing.
Required knowledge: 1. Programming skills (Python or Matlab) 2. Basic knowledge of ML and signal processing. Note: Neuroscience knowledge is not required.
Preferred knowledge: 1. Able to implement state-of-the-art ML algorithms and neural features. 2. Some prior knowledge on brain-machine interfaces.
Supervisor: MohammadAli Shaeri Xiaoyong Ni
Integrated circuit design for high-data-rate wireless telemetry in intracortical Brain-Machine Interfaces (BMIs)
The massive amount of neural data recorded by dense implantable electrode arrays must be transferred efficiently from the implant to an external hub. Current wireless telemetry systems have not yet fulfilled the data rates required for 1000s of recording channels. The trade-off between bandwidth, volume, and power consumption presents a significant challenge for circuit designers.
Students are expected to design and simulate RX front-end circuits like low-noise amplifiers in Cadence. Different modulation and demodulation methods will be explored for low-power and low-complexity circuit implementation. The student(s) will closely work with the advisor, PhD students, and postdocs in our group to optimize the wireless telemetry, design and test the prototype integrated circuits.
Objectives: 1. Understanding of the basic problems and research directions in wireless telemetry for implantable intracortical BMIs. 2. Learning to optimize the modulation and demodulation schemes. 3. Implementing, analyzing, and optimizing advanced analog and mixed-signal circuits.
Required knowledge: 1. Experience with analog/mixed-signal CMOS integrated circuit design in Cadence. 2. Familiar with wireless communication systems. 3. Experience with Matlab and HFSS is beneficial.
Preferred knowledge: 1. Able to implement analog/digital circuits on ASIC. 2. Some prior knowledge on neural interfaces.
Supervisor: Cong Ding
Circuit design for stimulation artifact cancellation in concurrent neural recording and stimulation
Concurrent neural recording and stimulation is required in various neurostimulation devices to monitor and treat brain disorders. However, the large stimulation artifacts caused by the large amplitude of stimulation pulses swamp the low-amplitude neural recordings and present a significant challenge for closed-loop stimulation therapy.
Students are expected to design artifact cancellation circuits in Cadence. Several candidate algorithms and their hardware architectures for efficient implementation will be explored and the circuit complexity will be assessed with Cadence simulations. The student(s) will closely work with the advisor, PhD students, and postdocs in our group to optimize the hardware and algorithm, evaluate it on real neural datasets, design and test the prototype integrated circuits.
Objectives: 1. Understanding of the basic problems and research directions in neural engineering. 2. Implementing and analyzing analog, digital and mixed-signal circuits. 3. Learning to optimize the hardware for low-power and compact implementation.
Required knowledge: 1. Experience with analog/mixed-signal or digital CMOS integrated circuit design in Cadence. Experience with Matlab and PCB design is beneficial but not necessary. Note: Neuroscience knowledge is not required.
Preferred knowledge: 1. Able to implement analog/digital circuits on ASIC. 2. Familiar with hardware design metrics. 3. Some prior knowledge on neural interfaces and neurostimulation.
Supervisor: Daney Alex Yiheng Fu
Neural interface circuit design using reinforcement learning
Reinforcement learning has shown promising results in automating integrated circuit design (e.g., for tuning circuit parameters and place-and-route optimization). In this project, jointly with LIONS lab, we will explore RL agents to design circuits for neural interfaces, where design tradeoffs (power, area, noise, latency, …) are particularly challenging. Our project will develop a new RL framework for building ASICs for modern neuroprosthetics.
Students are expected to design and simulate simple circuit blocks for neural interfaces such as low-noise neural amplifiers and neurostimulators in Cadence. Several candidate hardware architectures will be explored and used to train the RL tools developed by LIONS group and to study and quantify design tradeoffs. The student(s) will closely work with the advisor, PhD students, and postdocs in our group to optimize the designed integrated circuits.
Objectives: 1. Understanding of the basic problems and research directions in neural engineering. 2. Implementing and analyzing analog, digital and mixed-signal circuits. 3. Learning to optimize the hardware for low-power and compact implementation.
Required knowledge: 1. Experience with analog/mixed-signal or digital CMOS integrated circuit design (schematic and layout) in Cadence. Experience with Matlab and PCB design is beneficial but not necessary. Note: Neuroscience knowledge is not required.
Preferred knowledge: 1. Able to implement analog/digital circuits on ASIC. 2. Familiar with hardware design metrics. 3. Some prior knowledge on neural interfaces and neurostimulation.
Supervisor: Arshia Afzal
Resource-efficient machine learning algorithm design for on-implant neurological symptom detection
Recently, machine learning (ML) techniques have been used in early detection of neurological symptoms. However, although modern machine learning tools generally obtain a high accuracy in neural signal classification, their deployment on neural interface implants is severely limited due to stringent power, area, and latency constraints for these tiny implants. In this project, we aim to develop an efficient ML model to process neural data in real time, with low power consumption, small on-chip area and fast inference.
Students are expected to work with electrophysiological recordings such as scalp EEG, ECoG, and spikes (action potentials). Students will implement state-of-the-art ML models and develop ideas for their resource-efficient implementation, evaluate their performance on a given neural task, and estimate the hardware complexity. Students may focus on a particular metric (e.g., detection latency, power dissipation) or seek a reasonable trade-off between several performance metrics.
Objectives: 1. Understanding of the basic problems and research directions in neural engineering. 2. Applying machine learning algorithms to a real world application. 3. Gaining experience in biomedical signal processing. 4. Learning to optimize the designed algorithms for efficient hardware implementation.
Required knowledge: 1. Programming skills (Python or Matlab) 2. Basic knowledge of ML and signal processing. Note: Neuroscience and advanced hardware knowledge is not required.
Preferred knowledge: 1. Able to implement state-of-the-art ML algorithms. 2. Familiar with hardware design metrics. 3. Some prior knowledge on brain-machine interfaces.
Supervisor: Fei Tan MohammadAli Shaeri
Machine learning chip design for on-implant neural decoding
While modern machine learning shows a great potential in neural signal classification, its integration on neural interface implants is challenging due to severe constraints on power consumption and implant size. In this project, we aim to design an efficient feature extraction and/or ML processor to classify/decode neural data in real time, with low power consumption, small on-chip area and fast inference.
Students are expected to design candidate ML algorithms in Cadence. Various methods for efficient implementation will be explored (e.g., mixed-signal design, approximate computing, …) and the corresponding hardware complexity will be assessed with Cadence simulations. Students may focus on a particular metric (e.g., power dissipation, area, detection latency) or seek a reasonable trade-off between these hardware metrics. The student(s) will closely work with the advisor, PhD students, and postdocs in our group to co-design the hardware and algorithm, evaluate it on real neural datasets, design and test the prototype integrated circuits.
Objectives: 1. Understanding of the basic problems and research directions in neural engineering. 2. Implementing machine learning algorithms on chip. 3. Learning to optimize the hardware for low-power and compact implementation.
Required knowledge: 1. Experience with analog/mixed-signal or digital CMOS integrated circuit design in Cadence. Experience with machine learning algorithms and PCB design is beneficial but not necessary. Note: Neuroscience knowledge is not required.
Preferred knowledge: 1. Able to implement analog/digital circuits on ASIC. 2. Familiar with hardware design metrics. 3. Some prior knowledge on neural interfaces.
Supervisor: Huanshihong Deng Fei Tan MohammadAli Shaeri
Detecting mental fatigue and symptoms of psychiatric disorders from continuous neuronal recordings
Early detection of mental fatigue and changes in vigilance could be used to initiate neurostimulation to treat patients suffering from brain injury and various mental disorders. The goal of this project is to explore modern machine learning tools for detecting the onset of mental fatigue and symptoms of neuropsychiatric disorders, using multi-channel electrocorticography (ECoG) signals and local field potentials (LFP) recorded from primates (and possibly humans). The neural data is provided by our neurology/neurosurgery collaborator labs.
Students will study various correlating biomarkers (i.e., features) and implement state-of-the-art ML models to obtain a high classification performance and optimize the chosen algorithms in terms of detection accuracy and latency.
Objectives: 1. Understanding of the basic problems and research directions in neural engineering. 2. Applying machine learning algorithms to a real world (and challenging) application. 3. Gaining experience in biomedical signal processing.
Required knowledge: 1. Programming skills (Python or Matlab) 2. Basic knowledge of ML and signal processing. Note: Neuroscience knowledge is not required.
Preferred knowledge: 1. Able to implement and improve state-of-the-art ML algorithms. 2. Some knowledge on neural interfaces.
Supervisor: Jinhan Liu
Predicting epileptic seizures with diffusion processes
Research into seizures forecasting is a very important aspect for patients with epilepsy. Epilepsy affects 1% of the world’s population and seizures have so far been quite unpredictable. With the help of deep learning, early warning systems based on seizure forecasting are being developed. This project is about expanding a deep learning model for seizure forecasting based on diffusion models (based on the MG-TSD paper presented at ICLR 2024). The dataset will be an open access seizures dataset. At first, the student will familiarize itself with the dataset. Then the idea would be to adapt the network to this dataset and try to predict seizure as early as possible.
Objectives: 1. Understanding the current state of the art research in diffusion models and seizure forecasting. 2. Applying machine learning algorithms to a real world challenging application. 3. Gaining experience in machine learning applied to a biomedical application.
Required knowledge: 1. Good programming skills in Python 2. Good knowledge of DL, specifically diffusion models, RNNs and attention mechanisms. Note: Neuroscience knowledge is preferred but not required.
Preferred knowledge: 1. Prior projects on real data with DL models 2. Global knowledge of neurosciences
Supervisor: Amira Devika Neumann
Attention models for temporal decoding of neural data
Models used for sequence transduction have recently shifted from recurrent neural networks to attention mechanism. Such attention-based models (i.e. transformers) have achieved great success in the field of natural language processing. Neurological recordings (i.e., the electrical activity of the brain recorded by invasive or noninvasive electrodes) can be considered as multi-channel time sequences. While RNNs have shown promise in many tasks such as motor decoding, the attention mechanism is still unexplored in this context.
Students will apply attention-based models to process pre-recorded neural data. Students may need to explore the model structure to better fit into a specific neural task. Decoding performance will be compared with conventional approaches (e.g., LSTM). Limitations will be analyzed and solutions will be provided.
Objectives: 1. Understanding of the basic problems and research directions in neural engineering. 2. Applying machine learning algorithms to a real world (and challenging) application. 3. Gaining experience in biomedical signal processing.
Required knowledge: 1. Programming skills (Python or Matlab) 2. Basic knowledge of ML and signal processing. Note: Neuroscience knowledge is not required.
Preferred knowledge: 1. Able to implement and improve state-of-the-art ML algorithms. 2. Some prior knowledge on brain-machine interfaces.
Supervisor: Arshia Afzal MohammadAli Shaeri