skip to content

Department of Computer Science and Technology

Using body sound recordings to train computer models to predict Covid-19, and employing AI techniques in the quest to improve cancer treatment, were among the projects discussed at our Healthcare Research Showcase last week.  

The event - comprising a series of lightning talks by early-career researchers in this Department - was recorded and can be viewed below. Speakers covered the following topics: 

  • 04:56 Predicting outcomes for psychotic disorders, using brain connectivity and transcribed speech data: Dr Sarah Morgan;
  • 18:30 Designing a Robot Mindfulness Coach - a longitudinal study: Dr Indu Bodala;
  • 29:33 Automatic Covid-19 Diagnosis from Breathing, Cough Sounds, and Voice: Dr Jing Han;
  • 49:32 Integrative and interpretable AI methods and tools for bench-to-bedside research in cancer medicine: Dr Zohreh Shams, Dr Helena Andres-Terre, Dr Nikola Simidjievski.

This was the first event in a planned series of Research Showcase events we will be holding. The next will focus on some of our education-related research and will be held in spring 2021. For more details of either event, please contact our Research Strategy Manager, Helen Francis, by emailing research-strategy-manager@cst.cam.ac.uk

 

About the speakers
Accelerate Science Research Fellow Sarah Morgan focuses on applying data science approaches (including machine learning, network science and NLP methods) to better understand mental health conditions.

Postdoctoral researcher Jing Han has research interests in deep learning for human-centric multimodal affective computing, digital health, and intelligent signal processing.

Indu Bodala is currently working on longitudinal HRI studies where human users interact with robot wellbeing coach. Using an iterative design approach, she aims to develop autonomous robots that can deliver wellbeing interventions for long-term use.

In their talk, Zohreh Shams, Helena Andres-Terre and Nikola Simidjievski discuss how they are using an integrative and explainable bench-to-bedside pipeline to establish an AI based framework for personalised cancer medicine. Cancer research produces large amounts of complex heterogeneous data on different scales and from different sources. These three researchers talk about how integrative data analysis can provide better understanding of the underlying mechanisms of a biological process and ultimately lead to more accurate cancer diagnosis, prognosis and treatment planning.

 


Published by Rachel Gardner on Monday 1st February 2021