© Pint of Science, 2022. All rights reserved.
Artificial Intelligence for diagnosing dystonia
Dystonia is a term used to describe a family of tremors and contractions observed in patients that can seriously affect quality of life. Diagnosis is currently based on a subjective examination and can, therefore, be unreliable. This talk will introduce a computer-based approach to diagnosing dystonia that is quick, safe and easy to apply. Results presented will, for the first time, demonstrate the effectiveness of this technology to discriminate between two types of dystonia, organic and functional, which require very different treatments.
Modelling Parkinson's disease in the zebrafish
An animal model of Parkinson’s disease is essential for testing new therapies. Zebrafish are increasingly being used because they have similar neurons to those that die in people with Parkinson’s making them a good candidate for an animal model. Zebrafish copies of Parkinson’s related genes have been targeted for mutation to generate these models, which will then be used to train an evolutionary algorithm to classify the fish as mutants or wild-type based on swimming behaviour. The end goal is an effective model, that is successfully classified by an algorithm, for testing new therapies on.
Building the brain: how to make brain cells and wire them together
The brain is staggeringly complex. One of the wonders of human development is how approx 100 billion brain cells (neurons) are generated, migrate to the correct locations and form trillions of synaptic connections. It's not surprising that defects in these processes are responsible for a variety of neurological disorders, including autism and intellectual disability. I'll highlight some of the fascinating experiments that scientists have performed to study brain development, and describe some of our recent findings relating to an enzyme that we think is essential for the birth of neurons