Catherine Tong

Her achievements to date include building pioneering activity recognition models to advance the field of digital health, and contributing to the development of AlphaFold 3 to model the biomolecular world.

Whilst undertaking her PhD, Catherine Tong was captivated by the growing adoption of wearables and Internet of Things (IoT) devices. This inspired her to investigate continuous sensing and explore how these tools could evolve our understanding of physical and mental health.

Accurate monitoring of physical activity plays an important role in improving human understanding of physical health and in enhancing overall well-being. However, the field of ubiquitous sensing has long faced limitations due to the scarcity of labelled sensor data.

To address this challenge, Catherine and her collaborators built an open-source foundational model with significantly improved generalization ability and accuracy, based on a large-scale accelerometer dataset from the UK Biobank database. They successfully applied it to enhance digital monitoring research on clinical populations with motion disorders. This breakthrough AI model plays an important role in the field of limited labelled data, and can generalize to a series of downstream activity recognition datasets related to clinical and health applications.

Furthermore, recognizing the potential of multi-modal approaches, Catherine and her collaborators were pioneers in studying and extending their use in this area of wearable activity recognition. A key contribution to scientific research was their work on IMUTube, a groundbreaking system capable of translating motion information from the vision domain to the inertial domain. This innovation enabled the generation of vastly larger amounts of labelled sensor data, a crucial step in allowing deep learning models to generalize effectively.

Upon graduating with a PhD in 2022, inspired by the transformational release of the AI system AlphaFold 2, Catherine joined Isomorphic Labs, an Alphabet company pursuing AI-first drug design. In her role as Research Scientist, she is dedicated to fundamentally reimagining the process of drug discovery with the power and pace of artificial intelligence to ultimately advance human health.

In May 2024, Isomorphic Labs and Google DeepMind launched AlphaFold 3, which can predict the structure and interactions of all life’s molecules with unprecedented accuracy. As part of the Machine Learning team at Isomorphic Labs, Catherine is proud to have contributed to this groundbreaking release, which further highlights her dedication to leveraging cutting-edge technology to solve some of the most pressing challenges in healthcare and drug discovery.