Dr Sofya Titarenko

Profile

Throughout my school, university, research and academic career I was always interested in the application of mathematical skills to real-life problems. The medical field was always an attraction to me, however later I was also working with applied problems in geoscience, chemistry and engineering. Starting in the area of inverse and ill-posed problems (problems of reconstruction in Computerised Tomography) I later moved into the field of Data Science and Applied Statistics. I found that Statistics and Machine Learning tools give a new insight into the already known problems as well as allow to solve new ones.

Research interests

My main research interests lay in the development of mathematical algorithms for Applied Data Science problems. Special challenges present so-called Big Data problems, where mathematical algorithms have to balance accuracy, computational time, resources available and efficiency. A good example of computationally heavy problems is the analysis of genome data (e.g. sequence matching algorithms, searching for associations, etc.). With the ever-increasing volume of genomic data to process and analyse efficient algorithms become of paramount importance.

Computer vision is another topic of interest (Deep Learning in particular). Problems of classification/segmentation to assist with early diagnostics become very important in the medical field but can be also met in many other areas such as engineering, biology, etc. My up to date experience is with mammograms and fluorescence microscopy images. However, I am interested to expand it further.

My early career research falls in the area of ill-posed inverse problems. A combination of Machine Learning approaches and ill-posed inverse algorithms are something I am very keen to explore.

Research groups and institutes

• Statistics

Projects

<li><a href="//phd.leeds.ac.uk/project/1539-development-of-machine-learning-based-methodology-to-study-patterns-in-medical-data-(in-application-to-covid-19)">Development of machine learning-based methodology to study patterns in medical data (in application to COVID-19)</a></li> <li><a href="//phd.leeds.ac.uk/project/1361-fast-algorithms-for-efficient-biosequence-search">Fast algorithms for efficient biosequence search</a></li> <li><a href="//phd.leeds.ac.uk/project/1381-statistical-and-machine-learning-methods-to-predict-glucocorticoid-toxicity-outcomes-for-patients-with-immune-mediated-inflammatory-diseases">Statistical and Machine Learning Methods to Predict Glucocorticoid Toxicity Outcomes for Patients with Immune-Mediated Inflammatory Diseases</a></li>