The development of advanced machine learning and statistical analysis techniques has revolutionized our way of life, from handwriting recognition, image retrieval, weather forecasting and GPS route planning to online search engines and movie recommendation systems. It is now time to capitalize on successes in machine learning for biological discovery to benefit human health. Taking advantage of recent progress in machine learning and human genomics, this multidisciplinary project aims to combine deep learning architectures from machine learning and clinical medical images with the wealth of genetic information. The goal is to develop robust, general approaches for medical image analysis with an emphasis on capturing the architecture of the normal tissues in development and their changes in diseases, particularly cancer. The quantitative information extracted from the images will be used to assist pathologists in their decision-making and to analyse patient responses to therapy.
The candidate should be comfortable with programming, in particular in the C++, Python and Matlab languages, and be familiar with concepts in computer vision. Ideally he/she should possess a strong mathematics background and be motivated to work with biologists and clinicians to answer biological and clinical questions.
This is a high impact, unique PhD opportunity in which the candidate will receive exposure to and training in state-of-the-art techniques for machine learning and computer vision with opportunities to extend the algorithms for real-time applications. He/she will also work very closely with clinicians, pathologists and biologists within a highly interdisciplinary, collaborative, world-leading environment.
Supervisors: Prof. Jens Rittscher, Prof. Xin Lu