Application of Artificial Intelligence and Computer Vision to Analysis of Phenotypic Variation

Theme: Biodiversity & Ecology

Main Supervisor:

Norman MacLeod

Earth Sciences, NHM

Second Supervisor:

Diana Percy

Life Sciences, NHM

Project Description:

A variety of biological processes leave marks of their actions on phenotypes in the form of patterns of phenotypic variation. Traditionally, biologists have employed qualitative observation and simple measurements to describe and/or quantify these patterns to test process level hypotheses. However, recent developments in morphometrics, machine learning, and computer modelling have made a new generation of tools of unprecedented flexibility/sensitivity available to biologists for the purpose of finding, characterizing, comparing and interpreting these signals. The Natural History Museum is digitizing 1000s of slide-mounted specimens to facilitate the interpretation of phenotypic information. Data from these digital phenotypes will be analyzed and modelled using artificial intelligence, computer vision and computer modelling algorithms to address a wide range of contemporary biological/palaeobiological questions. An underpinning goal of this research programme is to reinvigorate and explore the role of phenotypic data in the assessment of environmental-phenotypic-taxonomic stationarity and stability.

Policy Impact of Research:

Artificial intelligence, computer vision, machine learning, and computer modelling techniques are all going to play major roles in 21st century science, including biology/palaeobiology. This research programme will equip students with unique skills that can be applied widely across the biological sciences in research, commercial, and policy development contexts.


Stay informed

Subscribe to our RSS newsletter by email.


Find Us

University College London is the administrative lead.

Pearson Building, UCL, Gower Street, London, WC1E 6BT

Follow us on Twitter