Madeleine Lowery

On the Human Health, Impact and Technology webinar series on 19 November 2024, Madeleine Lowery, Professor in the School of Electrical and Electronic Engineering and the Head of Subject for Biomedical Engineering at UCD spoke to host Professor Patricia Maguire about "Neuromuscular Systems and Neural Engineering". In case you missed it, here are the top takeaways and video recording.

From Engineering to Biomedical Innovation

Madeleine's passion for both engineering and medicine shaped her career path from an early age. Initially drawn to engineering, she chose electronic engineering due to her interest in its mathematical and technical aspects. However, she soon recognized a pathway into biomedical engineering, which led her to pursue a PhD in the field. She then moved to the United States, where she worked at Northwestern University and the Rehabilitation Institute of Chicago. There, she focused on upper limb prosthetics and targeted muscle reinnervation. After several years, Madeleine returned to UCD to continue her research on muscle function, modelling, electrical stimulation, and deep brain stimulation.

Interdisciplinary and Real world focused research

Madeleine specializes in neuromuscular systems and neural engineering, with a background in electronic engineering. Her research focuses on biomedical engineering, particularly the nervous and neuromuscular systems. Using advanced signal processing and computational modelling, she and her research group study how the nervous system controls muscle function and movement, and how these processes change with age or in the presence of diseases. The ultimate goal of her work is to improve motor function, enhancing movement for individuals. She is also exploring the development of technologies that can interface with the nervous system, such as electrical stimulation, to restore movement or alleviate symptoms in patients with various conditions.

Understanding Muscle Control Through EMG Signals

Madeleine's research explores how the nervous system controls muscle function in both healthy and diseased states. When a muscle contracts, it generates electrical signals similar to those produced by the heart (measured by an ECG) or the brain (measured by an EEG). These muscle-generated signals, known as electromyography (EMG), can be recorded either through sensors implanted in the muscle or placed on the skin's surface. EMG signals contain valuable information about the muscle’s state and the nerves that control it. By applying advanced signal processing and engineering techniques, Madeleine’s team extracts and analyses this data to better understand how individual nerves control muscle fibres, enabling movement and interaction with our environment. This combination of experimental methods and signal processing provides a unique window into the workings of the nervous system.

Deep brain stimulation

Madeleine’s research focuses on deep brain stimulation (DBS), a therapy that has been used to treat the symptoms of Parkinson’s disease over the past 30 years, helping almost 300,000 people with neurological conditions during that time. DBS involves implanting electrodes into the brain to deliver electrical pulses that modify neuronal activity. Her team uses engineering models to study how these pulses affect neuron firing and muscle activation patterns, with the goal of improving current treatments and developing advanced technologies, such as closed-loop stimulation that can automatically adjust the stimulation delivered in response to patient symptoms and side-effects. Since 2021, DBS has been available in Ireland, thanks to neurosurgeon Ms Catherine Moran at Beaumont Hospital, who along with neurologist Professor Richard Walsh introduced the therapy locally, reducing the need for Irish patients to travel abroad for treatment.

 

Advancing Brain-Computer Interfaces for Motor Function Restoration

Brain-computer interfaces (BCIs) offer great potential for restoring motor function in individuals with severe neurological impairments, such as those with spinal cord injuries or brainstem strokes. These systems aim to interpret brain signals, enabling patients to interact with their environment through speech prostheses, cursor movement, or robotic arms.

There are two main approaches: implantable sensors, which are placed directly on the brain's cortex, and non-invasive sensors, such as EEG.  Madeleine's team is focused on improving implantable sensors by developing new electrodes that better integrate with brain tissue, though challenges remain around their long-term stability. Despite these obstacles, significant global progress is being made in advancing BCIs for speech and motor function restoration.

The Role of Humans in Brain-Computer Interface Systems

The vast amount of data generated by brain-computer interfaces and related technologies presents significant challenges, particularly in terms of interpretation, management, and extracting meaningful insights. One crucial aspect of this process is the "human in the loop." While machine learning algorithms play a key role in analysing and processing the data, the human element is equally important. Patients or users must learn how to adjust and fine-tune their signals to work effectively with the system, essentially collaborating with the machine to close the loop and achieve the desired outcomes. This interplay between human input and machine learning is essential for the success of these technologies.

Optimizing Deep Brain Stimulation for Parkinson's Treatment

DBS is highly effective in treating Parkinson’s disease symptoms, such as tremors and slowness of movement, with improvements of up to 60-80%. While the devices are made by large medical companies, Madeleine’s team focuses on understanding and optimizing how they work. One challenge is adjusting the wide range of stimulation parameters—such as amplitude, duration, and frequency—to each patient’s needs. To improve this, her research is developing closed-loop algorithms that can automatically monitor symptoms and adjust stimulation settings in real time, offering more personalized and effective treatment

Advancing Closed-Loop DBS and Cortical Stimulation for Parkinson's

Patients with DBS still rely on Levodopa for non-motor symptoms, but a closed-loop system being developed by Madeleine’s team could automatically adjust stimulation in real time based on changes in symptoms, offering more precise control. This system would adapt to fluctuations throughout the day, and even long-term changes. While cortical brain stimulation shows promise as a potential alternative to DBS, research is still needed to determine efficacy and to identify the optimal areas to stimulate and the correct levels of stimulation. Studies suggest that adapting stimulation to the body's natural rhythms, such as during sleep, could improve symptom management and provide more refined control.

Advancing Rehabilitation and Neurological Recovery

Understanding the most effective rehabilitation therapies for patients, particularly those with Parkinson’s disease, is a key area of research. A few years ago, Madeleine's team conducted a study on LSVT BIG therapy, an exercise-based intervention focusing on large movements, which has been shown to improve certain motor symptoms in Parkinson's patients. Collaborating with occupational therapists at the Royal Hospital in Donnybrook, they used sensors to track changes in movement and gait speed in patients undergoing the therapy. Beyond exercise interventions, research in the field of neural engineering and rehabilitation extends to the development of advanced technologies, including prosthetics with multiple degrees of freedom for more natural movement and sensory feedback, as well as rehabilitation robotics for patients recovering from brain or spinal cord injuries.

Future

In the next decade, Madeleine hopes to see deep brain stimulation (DBS) and other therapies reach a wider patient population, offering benefits earlier in the disease process with more personalized interventions. She envisions combining sensing technologies with stimulation or rehabilitation therapies to create highly tailored, real-time treatments. By closing the loop between sensing and intervention, this approach could lead to more effective, individualized care, improving outcomes for patients with neurological conditions.