- Course: Computer Vision and Deep Learning PhD
- PhD title: Machine Learning for Glaucoma Assessment using Fundus Images
- Year of graduation: 2019
- Nationality: Colombian
- Job title: Postdoctoral Research Fellow
- Company: University of Leeds
Andres Diaz-Pinto is a Postdoctoral Research Fellow specialising in Machine Learning and Deep Learning. Based in the Centre for Computational Imaging and Simulation Technologies in Biomedicine (CISTIB), he leads the Artificial Intelligence (AI) team, which develops AI technologies with healthcare applications. After finishing his PhD at Universitat Politècnica de València in Spain, Andres was offered a postdoctoral position at the University of Leeds.
“As ‘Andrew Ng’ says, AI is going to be the new electricity. This is an exciting time to be a machine learning and deep learning researcher,” said Andres. “There are fascinating applications in which AI will change the way we interact with the world, and I want to continue doing cutting-edge research to make it possible.”
AI is going to be the new electricity. This is an exciting time to be a machine learning and deep learning researcher.
Tackling diabetes and cardiovascular diseases
Healthcare applications of the research includes conceptualising machine learning systems which can aid with assessing medical conditions such as diabetes and cardiovascular diseases. Andres uses a variety of highly complex machine and deep learning techniques to carry out the research.
He said: “I develop advanced AI tools to analyse different medical image modalities and genomic information in the UK Biobank cohort. As part of my research, I conduct a ‘multi-modality image analysis,’ which is achieved using deep learning. The aim is to develop a machine learning system that will be able to improve technologies used by medical professionals to assess both diabetes and cardiovascular diseases.
“Machine learning is a subfield of AI that involves a computer using algorithms, which continually improve as it receives more data, to ‘learn’ based on experience.”
Andres continued: “Through the image analysis, I can study different types of ‘omic’ information, which relates to clusters of molecules, to identify correlations between the genetics of an organism and its observable properties in an environment – through a method known as ‘deep phenotyping.’ Using AI techniques, it is possible to analyse this information on a large scale.
“Using deep learning techniques, I am able to develop and refine 3D cardiac shape reconstruction methods to analyse cardiovascular systems in a way that is extremely detailed. Statistical learning methods are also used when developing ‘deep cardiac phenotyping’ methods in imaging genetics.”
Using deep learning techniques, I am able to develop and refine 3D cardiac shape reconstruction methods to analyse cardiovascular systems in a way that is extremely detailed.
He added: “The purpose is to examine the genetic bases of these cardiac phenotypes, which have been segmented from cardiac magnetic resonance (CMR) images, to assess the likelihood of cardiovascular disease and possible interventions.”
Leading a team of AI researchers
Andres leads a team of researchers, guiding and coaching three researchers working on PhDs, and six MSc students. He enjoys supporting and inspiring others to generate research impact that could make a positive difference in society.
He said: “In our group, most of our time is spent on discussing the latest advances in computer vision and machine learning techniques,” said Andres. “In addition to this, writing, programming and creating software prototypes are also essential to carry out our research.”
Collaborations in the multidisciplinary group enable the researchers to learn from each other and develop new ideas, Andres explained.
He added: “What motivates me the most is the possibility of making a significant and positive impact on people's lives by using machine learning. As an AI team leader, I always seek ways of creating more advance and powerful machine learning tools to solve complex problems and, at the same time, foster the professional development of my group members.”
What motivates me the most is the possibility of making a significant and positive impact on people's lives by using machine learning.
Find out more
To learn more about the Centre for Computational Imaging and Simulation Technologies in Biomedicine (CISTIB), and being a part of Leeds’ postgraduate community, visit our CISTIB and research degrees pages.