Imperial engineers awarded grants to develop commercial and societal research

Imperial researchers from have each been awarded a lump sum of €150,000 as part of the ERC Proof of Concept Grant scheme. The grants provide support for previous and current ERC grantees who want to explore the commercial or societal potential of their work.

Decoding nerve signals to control assistive devices

Professor Dario Farina from our Department of Bioengineering leads the AxonCtrl project.

During previous projects, the team developed intramuscular micro-electrode arrays that can be inserted through the skin without surgery to record muscle electrical activity. The activity of of individual spinal motor neurons is identified with AI. 

This is the first minimally invasive interface that can access large populations of human spinal cord neurons at the single-cell level. In this new project, these electrodes will be implanted in paralysed muscles of people with spinal cord injury to capture residual voluntary motor signals. In this way, the interface will provide a new way to decode patients’ movement intentions for controlling assistive technologies.

Professor Farina said: “This project will turn our lab neural interface into a wearable system that lets people with tetraplegia use their own nerve signals to control assistive devices more naturally in everyday life.”

Revolutionising nature cinematography

Capturing nature’s most fleeting moments, such as a lemur leaping or dragonflies dueling over water has always relied on a mix of skill and luck. It also demands extensive setup and long waits in harsh environments. 

The Dynamic Automatic Region Tracking (DART) project, led by Dr Huai-Ti Lin, Department of Bioengineering, aims to make these moments become routine and much easier to capture. 

DART revolutionises nature cinematography with bioinspired algorithms and cutting-edge hardware; using motorized mirrors to steer a camera’s view, it enables ultra-fast, high-resolution tracking of unpredictable subjects. The system integrates two cameras: a wide-angle search camera to detect action and a telephoto homing camera to capture it in stunning detail. Its bioinspired vision algorithm and controller enable low-latency, data-efficient, and energy-efficient continuous operation. 

This project will refine DART’s tracking algorithms and hardware controls to create a modular, plug-and-play solution for applications in agri-tech, environmental monitoring, industrial automation, entertainment, and wildlife documentary production. Building on our strong academic expertise in robotics and biomechanics, DART is now advancing toward real-world demonstrations with early partners. By capturing what once seemed unfilmable, DART opens new frontiers in motion analysis and visual storytelling.

Dr Lin said: “Too many creative research instruments remain confined to the laboratory. The ERC Proof of Concept grant enables us to take our technology into real-world settings and accelerate its path to impact. It is pivotal to completing my lab’s roadmap from fundamental science to engineering innovation and industry application.”

Pre-natal heart screening

Professor Bernhard Kainz, Department of Computing, has been awarded an ERC Proof of Concept (PoC) Grant to advance the early detection of congenital heart disease (CHD) through artificial intelligence.

The project, titled CHARMS (Congenital Heart Anomaly Representation and Machine-learning for Screening), will build on Professor Kainz’s ERC Consolidator GrantMIA-NORMAL, which pioneers new algorithms for normative representation learning in medical imaging. These methods enable machines to learn what healthy anatomy looks like and then flag subtle abnormalities as potential disease indicators.

Congenital heart disease is the most common and deadliest congenital malformation, affecting about 1% of pregnancies, yet current ultrasound screening programmes miss around half of severe cases. Early detection significantly improves outcomes, but detection rates vary widely depending on sonographer expertise and local resources.

CHARMS will take this foundational research into clinical practice together with clinical partners, focusing on ultrasound-based prenatal screening. By analysing entire ultrasound video sequences instead of selected still images, the project will develop robust anomaly detectors capable of identifying both known and previously unseen heart malformations. Generative AI methods will provide interpretable "what-if" scenarios for clinicians, showing how a scan would appear if the foetus were healthy or exhibited clear signs of a specific condition.

Working in partnership with King's College London and Fraiya Ldt., a spin-out company co-founded by Professor Kainz, CHARMS will test its prototypes with sonographers and fetal cardiologists at leading hospitals. The aim is to ensure usability, trustworthiness, and compliance with the EU AI Act, creating a pathway towards adoption in routine prenatal care.

Professor Kainz said: "Most babies are healthy, and it is crucial that we give clinicians tools to confirm this confidently while catching the rare but serious cases of congenital heart disease as early as possible. With CHARMS we want to democratise access to expert-level prenatal screening, making outcomes less dependent on where or by whom a scan is performed."

If successful, CHARMS could reduce neonatal deaths from CHD in Europe by up to 20% and cut unnecessary referrals, saving healthcare systems millions of pounds annually. The technology also has the potential to open new markets in community and even home-based ultrasound screening, supported by smartphone-operated devices.

Non-invasive and rapid cancer surveillance

Dr Jia Li, Department of Metabolism, Digestion and Reproduction, has received a grant for her project ‘Risk stratification and early diagnosis of bowel cancer in people with Lynch syndrome (RED-LYNCH)’

Lynch syndrome (LS) is a hereditary condition that predisposes carriers to various cancers, particularly colorectal cancer (CRC). Approximately 40% of individuals with LS develop colorectal tumours before the age of 40, often with faster disease progression.

Due to this elevated lifetime risk of CRC, clinical guidelines recommend colonoscopies every other year starting at age 25. However, colonoscopy is invasive, costly, and time-consuming, creating a significant burden for both patients and healthcare systems. 

Furthermore, it does not provide molecular-level data, which have significant potential to improve CRC risk stratification and inform the development of personalised prevention strategies. 

This project aims to develop a non-invasive and rapid surveillance tool based on circulating metabolite biomarkers aiming to enable CRC risk stratification and monitoring over time. 

Additionally, the tool can provide metabolic profiles that may inform personalised nutritional interventions to reduce CRC risk in LS carriers. 

Dr Lia said: "What’s exciting about this ERC PoC grant is that it helps us move our research beyond the lab and closer to real-world impact. We aim to develop a tool that can more precisely assess cancer risk and support earlier detection, ultimately contributing to more personalised cancer prevention.” 

Coupling energy harvesting and storage

The need to power batteries in remote areas off-grid as well as the rapid development of the Internet-of-Things (IoT) sensor networks, which rely on rechargeable batteries to operate, are increasing the demand to find new solutions to couple energy harvesting systems with energy storage. 

SolBat, a proof-of-concept project led by Professor Cecilia Mattevi, Department of Materials, aims to develop a light-rechargeable battery to power a wide range of IoT sensors. 

The battery is miniaturized and rechargeable upon light illumination (indoor and outdoor) without an external voltage or current applied, enabling a reliable real-time continuous sensing and off-grid energy autonomy. 

Professor Mattevi said: “This technology tackles the growing need for off-grid, continuous IoT sensing by unifying harvesting and storage in one light-rechargeable battery. The proposed technology addresses a real and growing need with both societal and commercial dimensions and the absence of commercial solutions with equivalent functionality. The potential applications could span from health monitoring to environmental sensing.”