Acoustic tweezers are devices that allow non-contact manipulation of small objects in fluids. They have applications ranging from cell science and tissue engineering to developing new 3D displays. This project will use artificial intelligence to optimise the operation of these devices and improve our understanding of how the devices work in practice.
Electronically-controlled acoustic tweezers use arrays of ultrasound transducers to produce carefully designed ultrasonic fields which act to force objects to the desired locations. By changing the applied signals, and hence the fields, the objects can be moved. For simple systems the required signals can be predicted analytically. Additional complexity, due to practical factors (such as local acoustic resonances or transducer variations) or more difficult requirements (multiple trapped objects, complicated patterns), increases the difficulty of these analytical approaches and ultimately limits what can be achieved.
Of all the possible applications of acoustic tweezers, biological cell manipulation is one of the most challenging technically, but also one with the potentially greatest impact in improving healthcare outcomes. Machine learning offers an opportunity to overcome a roadblock in the progress of acoustic tweezers from an engineering puzzle to a tool for biologists and eventually medical researchers. This potential for progress comes with an obligation to consider the safety and ethics of where these devices will end up, balanced by a clearer understanding of their potential capabilities.
Machine learning has been widely applied to scientific and technical problems, but much of the research uses an algorithm as a black-box solution: solving the immediate problem without providing new understanding of the system. This limits the potential impact of the solution and also raises questions about who is responsible for the direction of inquiry taken, the degree to which results can be trusted and ultimately who is responsible for the safety of the system. Here we will investigate transparent approaches that allow us to solve the immediate control problem in an intelligible manner and so let us better understand the underlying system behaviour and improve our designs, while retaining the researcher as central to the decision making process.
In this PhD you will design computational models of acoustic tweezers and then apply machine learning to develop new control strategies. Relatively simple devices will be used to validate the results experimentally. The resulting control strategies will be studied to probe the underlying physical processes that lead to their success and this new knowledge used to improve the design and application of acoustic tweezers. Inverting the previous approach (where people used knowledge of the physical system to develop control strategies) offers an exciting route to new insights. By developing a full picture of the potential benefits of these devices we will contribute to the evidence base for moving them along the pathway towards biological, therapeutic and clinical applications.
This project is associated with the UKRI CDT in Accountable, Responsible and Transparent AI (ART-AI), which is looking for its second cohort of at least 10 students to start in September 2020. Further details can be found at: http://www.bath.ac.uk/centres-for-doctoral-training/ukri-centre-for-doctoral-training-in-accountable-responsible-and-transparent-ai/
Desirable qualities in candidates include intellectual curiosity, a strong background in maths and programming experience.
Applicants should hold, or expect to receive, a First or Upper Second Class Honours degree. A master’s level qualification would also be advantageous.
Informal enquiries about the research should be directed to Dr Charles Courtney: [email protected]
Enquiries about the application process should be sent to [email protected]
Formal applications should be made via the University of Bath’s online application form: https://samis.bath.ac.uk/urd/sits.urd/run/siw_ipp_lgn.login?process=siw_ipp_app&code1=RDUCM-FP02&code2=0002
Start date: 28 September 2020
ART-AI CDT studentships are available on a competition basis for UK and EU students for up to 4 years. Funding will cover UK/EU tuition fees as well as providing maintenance at the UKRI doctoral stipend rate (£15,009 per annum in 2019/20, increased annually in line with the GDP deflator) and a training support fee of £1,000 per annum.
We also welcome all-year-round applications from self-funded candidates and candidates who can source their own funding.