Machine Learning Methodologies for Personalized Health

Supervisor:   Neil D. Lawrence   

It is now possible to have several perspectives on a patient. mHealth provides information derived from mobile phones. Full genotyping of patients is becoming affordable, providing information about genetic background. The phenotype of disease is becoming better characterised than ever before. Techniques such as transcriptome analysis allow a highly detailed characterization of the state of a tissue. Finally the UK government’s midata initiative (and similar initiatives elsewhere) may eventually allow patients to provide information about their consumer spending habits as well as social network behaviour.\ \ These different data modalities need to be combined into one model of the patients well being. There are major challenges with doing this: models need to be applied across millions of patients and for any given patient many information modalities will be missing. Addressing data of this type requires new machine learning methodologies. This project will focus on combining data from different modalities within the same probabilistic model.

Inferring Complex Hidden Causes of Disease through Probabilistic Models

Supervisor:   Neil D. Lawrence   

The increased ability to measure individuals’ genetic profiles ( through single nucleotide polymorphisms), combined with the ability to characterize a disease activity through gene expression, and other biomarkers, should reveal more realistically complex webs of causal factors for disease. Understanding these causal factors would enable personalised interventions targeted to an individual’s genetic, environmental (including concurrent treatments) and treatment preference profile.

This vision is challenging on two fronts: first, integration of different sources of (genomic) biological data is not straightforward; second, environments are not artificially controlled. In practice, disease occurrence and progression is often triggered by a combination of genetic and environmental factors. Environmental factors need to be assimilated with the genetic background (through the genomic data) and placed in a unified modelling framework to characterize the disease.

The objective of this project is to address these issues. Our aim is to perform statistical inference from these models to enable us to resolve the determinants of a given disease and its responses to treatments. The research will involve amalgamation of several different research areas, covering health, biology, computational and statistical inference.

The project will involve a large amount of mathematics, in particular probability theory and advanced linear algebra.

Data Driven Mechanistic Models for Systems Biology

Supervisor:   Neil D. Lawrence   

Models of biological systems, such as gene networks, are useful in extracting meaning from quantitative data obtained from specific biological systems. In this PhD proposal we are interested in how things actually work in real observable biological systems. To do this we will make use of machine learning to infer mechanistic models biological systems.

Biological systems are immensely complex, and extracting all the data necessary to characterize the system is often impossible. It is therefore important to exploit other sources of information when modelling biological systems. Once such source of information is “mechanistic models”. These are models of the underlying physical properties of the system. In this project we will ensure that such physical models can be easily combined with data driven machine learning approaches, aiming to obtain the best of both worlds: mechanistic modelling and data driven machine learning models.

The project will involve a large amount of mathematics, in particular advanced linear algebra and calculus.

Animation by Machine Learning with Motion Capture Data

Supervisor:   Neil D. Lawrence   

This project is about using Machine Learning to model human motion for animation, with a particular focus on the demands of computer games. The idea is to learn what natural motion looks like, and then combine it with constraints to develop an animation sequence. The constraints could be animator imposed, or imposed by the computer game. A typical scenario might be that the player’s character is required to interact with a character in the game, for example a player might be given an object in the game. The constraint could be that the hands of the player touch the hands of the character giving the object. With the current approach to animation (looking up a library of motion capture data) such a constraint is very difficult to fulfill as the required motion won’t exactly match a sequence in the library. By modelling natural motion through machine learning, we should be able to generate a new sequence to satisfy the constraint.

The models of motion will be developed using Gaussian processes. In particular the project will make use of the “Gaussian Process Latent Variable Model” (Lawrence, 2003) which has already shown a lot of promise in this domain, and “Latent Force Models” (Alvarez et al, 2009) a recently developed approach to learning based on physics and probabilistic models.

The project will involve a large amount of mathematics, in particular advanced linear algebra and calculus.

subscribe via RSS