PhD Studentship - AniBrain: Advancing Non-Invasive Brain-Machine Interfaces Through Large Language Models

Number of positions: 1

This 3.5 year PhD project is fully funded by the Strategic Doctoral Landscape Award (sDLA) Scholarships and is available to home students only. Tuition fees will be paid and the successful candidate will receive an annual tax free stipend set at the UKVI rate (£19,237 for 2024/25).

Interested applicants are encouraged to send their CV and cover letter to jingyuan.sun@manchester.ac.uk at their earliest convenience. If we find a suitable candidate, we might end this recruitment earlier than the advertised deadline.

Non-invasive brain-machine interfaces (BMIs) present an exciting frontier for helping individuals to communicate more effectively, especially those with language disabilities. While invasive BMIs have achieved impressive results in restoring or augmenting speech, they raise significant safety concerns due to the surgical procedures required. Conversely, non-invasive BMIs, which use external sensors such as MEG, EEG or fMRI, offer a safer and more accessible approach for a broader range of potential users.

This project aims to develop and refine non-invasive BMIs that integrate state-of-the-art AI-based neural signal processing techniques and large language models (LLMs). By leveraging the powerful linguistic representations of LLMs, we seek to decode and interpret neural activity related to language in a way that is both accurate and robust. Through careful methological design with advanced representation learning and transfer learning technologies, we will optimize the interface to handle individual variability and minimize environmental noise, thereby enhancing communication for those with language impairments. Ultimately, this research will contribute to more accessible, user-friendly, and reliable BMI systems, opening up transformative communication possibilities for individuals with language disabilities and, in the long term, extending these innovations to the wider population seeking safe, non-surgical solutions.

Pursuing a PhD is a journey filled with opportunities and challenges, requiring both academic potential and strong personal qualities. We hope candidates possess the following attributes:

Personal Qualities

Self-discipline: Ability to plan time effectively, maintain research continuity and efficiency, and demonstrate strong self-management, motivation, and goal-setting skills.

Diligence: A willingness to take on complex, long-term research tasks with resilience and commitment, investing time and energy consistently.

Passion: Genuine interest and enthusiasm for academic research, with a strong drive to explore and solve scientific challenges creatively.

Fluent English Proficiency in Listening, Speaking, Reading, and Writing

Candidates should ideally have an IELTS average score of 6.5 or higher (or equivalent), as proficiency in English will be highly beneficial for studying and living in the UK. Academic writing experience in English is a plus.

Academic Background and Technical Skills

Programming experience in artificial neural networks, deep learning, large models, natural language processing, and computer vision is advantageous.

Proficiency in common machine learning frameworks (e.g., PyTorch, TensorFlow, MxNet, Huggingface-Transformer) is highly desirable.

Not required but beneficial: background knowledge in cognitive neuroscience, medical imaging, or bioinformatics.

Interested applicants are encouraged to send their CV and cover letter to jingyuan.sun@manchester.ac.uk at their earliest convenience. If we find a suitable candidate, we might end this recruitment earlier than the advertised deadline.

Advert information