CHINA SCHOLARSHIP COUNCIL: Application of machine learning to the analysis of spectroscopy data in neurological disease

Details

Machine learning and advanced data analysis techniques are poised to make a significant contribution to healthcare in the future. Alongside this, biophotonic technologies, capable of providing highly detailed molecular information, are entering clinical testing. Combining biophotonic technology with advanced data science and machine learning has great potential to lead to a step change in disease diagnosis and monitoring.

The biophotonics technology derived data is high dimensional with very little a priori knowledge of the biomarkers that are relevant to disease diagnosis and prognosis. From a clinical perspective, it is also important that any analysis method has the capability to provide insight into potential biological mechanisms. This leads to a need for careful formulation of the machine learning problem and the ability to interpret spectral information, identification of clusters as well as relating these to specific protein conformations. These demand consideration of both unsupervised and supervised machine learning methods.

In addition to the high dimensionality of the data, clinical data consists of significant patient variability. Providing uncertainty quantification of any analysis is therefore of utmost importance, particularly when they are used in clinical decision-making. The choice of machine learning and/or advanced statistical data science methods should have the capacity to provide measures of prediction uncertainty as part of the modelling process.

Once the machine learning methodology is demonstrated successfully on the biophotonic data, there will be a need to develop a tool that can be used in a clinical decision-making setting. The student will work alongside leading data scientists and clinicians to develop an advanced data analysis platform, embedding the successful algorithms within it. The choice for the software platform for the tool development can be in Matlab, Python and/or R.

Entry Requirements:

Candidates must have a first or upper second class honours degree or masters degree in an subject with mathematical/programming training e.g. mathematics, physics, chemistry, engineering, computer science. Candidates with a life science background may be considered but will need to demonstrate evidence of high level mathematical/programming skills.

To be eligible for a CSC scholarship, you must:

• be applying to start a new full-time PhD programme at Sheffield
• have a strong academic background
• be a national of and reside in, mainland China (not including Hong Kong or Macau) – see CSC policies
• intend to return to China once your programme is completed
• hold a study offer from the University of Sheffield (see below for further details)
• satisfy all of the CSC's selection criteria 

How to apply:

Please complete a University Postgraduate Research Application form available here: https://www.sheffield.ac.uk/postgraduate/phd/apply/applying

Please clearly state the prospective main supervisor in the respective box and select ‘School of Medicine & Population Health: Neuroscience’ as the department.

Proposed start date - October 2024

For further information on the CSC scheme please see - https://www.sheffield.ac.uk/postgraduate/phd/scholarships/csc